# growthepie — Full Content > growthepie is the open analytics platform for the Ethereum ecosystem, > tracking real-time and historical metrics across Ethereum Mainnet, Layer 2 > networks, and onchain applications. This file is the companion to > /llms.txt and contains the full prose of the /answers section — direct, > data-backed answers to common questions about Ethereum and its wider > ecosystem. Recomputed daily from growthepie's public API. > Data licensed CC BY-NC 4.0. Source: https://www.growthepie.com ## What is the most popular Data Availability (DA) layer for Ethereum L2s? URL: https://www.growthepie.com/answers/most-popular-da-layer-for-ethereum-l2s ### Direct answer The most popular Data Availability (DA) layer for Ethereum L2s is **Ethereum DA (blobs)**, used by 15 of 25 growthepie-tracked L2s including Arbitrum One, Base Chain, Ink. Other DA layers in use: Celestia (3), EigenDA (3), Other / Custom DA (3). Data: 2026-05-27 UTC. Live tracker: growthepie.com/data-availability. ### Full answer Short answer (data 2026-05-27 UTC): The most popular DA layer for Ethereum L2s is Ethereum DA (blobs) (15 L2s — e.g. Arbitrum One, Base Chain, Ink). The full split across 25 tracked L2s ranks as: 1. Ethereum DA (blobs): 15 L2s (Arbitrum One, Base Chain, Ink and 12 more); 2. Celestia: 3 L2s (Gravity, Manta Pacific, Plume Network); 3. EigenDA: 3 L2s (Celo, Mantle, MegaETH); 4. Other / Custom DA: 3 L2s (Arbitrum Nova, Fraxtal, Metis); 5. Ethereum DA (calldata): 1 L2s (Loopring). By USD fees paid over the last 30 days the leader is unavailable. Updated daily — adoption counts, fee totals, and per-MB costs on this page are recomputed from growthepie's public API every day. Adoption is a snapshot at 2026-05-27 UTC; the underlying L2 → DA mapping changes whenever a chain switches DA provider or growthepie adds new coverage. Adoption ranking — how many L2s use each DA layer As of 2026-05-27 UTC, 25 tracked Ethereum L2s split across DA layers as follows: Ethereum DA (blobs): 15 L2s; Celestia: 3 L2s; EigenDA: 3 L2s; Other / Custom DA: 3 L2s; Ethereum DA (calldata): 1 L2s. Why the ranking matters The "most popular DA layer" question has two equally valid answers, and this page reports both: - By number of L2s — this is "which DA do chains choose?" The vast majority of L2s today post data to Ethereum mainnet via blobs because it gives the strongest possible security guarantee (Ethereum's validators, Ethereum's consensus, Ethereum's economic backing). - By USD fees paid — this is "which DA captures the most economic activity?" Ethereum DA typically leads here too because the L2s using it have the largest aggregate throughput — even when alt-DA is much cheaper per byte, the Ethereum-backed L2s post far more bytes. Where the two rankings can diverge is the per-megabyte cost: Celestia and EigenDA are typically much cheaper per byte than Ethereum DA. Cheap-per-byte is the entire economic motivation for any chain to consider alt-DA; the question is whether the cost saving outweighs the additional trust assumption (the alt-DA layer's own validators or restakers). DA layers in detail Ethereum DA (blobs). 15 tracked L2s use it. Since Dencun (March 2024, EIP-4844) Ethereum mainnet provides DA through blobs — large 128 KB data attachments priced separately from regular L1 gas. Fusaka (December 2025) added PeerDAS (EIP-7594) which lets nodes verify blob availability via sampling, enabling further capacity increases. Chains using it: Arbitrum One, Base Chain, Ink, Linea, Lisk, Mode Network, OP Mainnet, Ronin, Scroll, Soneium, Starknet, Taiko Alethia, Unichain, World Chain, ZKsync Era. Celestia. 3 tracked L2s use it. Standalone PoS blockchain dedicated to DA, mainnet October 2023. 30-day average per-megabyte cost: $0.0351. Chains using it: Gravity, Manta Pacific, Plume Network. EigenDA. 3 tracked L2s use it. Built on EigenLayer — Ethereum's restaking ecosystem. 30-day average per-megabyte cost: $0.0464. Chains using it: Celo, Mantle, MegaETH. Avail. 0 tracked L2s use it. Standalone modular DA layer with KZG commitments and data availability sampling. Chains using it: none. Some chains use other or custom DA approaches — count: 3. Chains: Arbitrum Nova, Fraxtal, Metis. USD fees paid to each DA layer Total USD fees paid by L2s to each DA layer over the last 30 days (data 2026-05-27 UTC): - Ethereum DA (blobs) — unavailable - Celestia — unavailable - EigenDA — unavailable All-time cumulative USD fees: - Ethereum DA (blobs) — unavailable - Celestia — unavailable - EigenDA — unavailable Avail is not yet exposed in growthepie's fees endpoint, so its dollar totals are unavailable on this page. Live charts: growthepie.com/data-availability. What "most popular" misses — the trade-off behind each chain's choice Adoption counts alone don't tell you why a chain picked the DA it picked. The honest framing: - DeFi, settlement, treasury-grade L2s tend to choose Ethereum DA. The premium is worth the strongest possible security; DeFi users are sensitive to trust assumptions. - Gaming, social, NFT, high-throughput consumer L2s are more likely to use alt-DA. The cost difference matters at scale and the typical use case can tolerate the extra trust assumption. - Hybrid chains split the difference, posting some data to Ethereum and bulk to alt-DA. The dominance of Ethereum DA in the adoption count therefore says less about Ethereum DA "winning" and more about most current L2s being settlement-leaning rather than throughput-leaning. As consumer L2s scale, the alt-DA share is the metric to watch. Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue and filter to L2s the same way the rest of growthepie's answer pages do (bucket !== "Layer 1", deployment === "PROD", excludes sidechains and aggregate keys). 2. Read the string field from each L2's master entry, normalise it to one of the canonical buckets (Ethereum blobs / Celestia / EigenDA / Avail / other), and count. 3. Pull for the daily USD-fees-paid series per DA layer. Sum into daily / 30d / 90d / all-time. 4. Pull for per-megabyte cost. Take the 30-day average of the daily USD column. 5. Sort the buckets descending by L2 count for the headline ranking; secondary rankings (USD fees over 30 days, $/MB) are reported alongside. All values shown were generated on 2026-05-27 UTC. Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Some supporters operate chains using specific DA layers. The ranking on this page is computed mechanically from — chains and DA layers do not influence inclusion or placement, and supporters do not receive ranking adjustments. Full list of supporters: growthepie.com/donate. Cross-check this answer. Independent DA-layer data sources you can compare against include L2BEAT's DA risk view (per-chain DA classification and risk profile), the DA layers' own dashboards (celestia.org, eigenda.xyz, availproject.org), and the published EIPs themselves (EIP-4844 for blobs, EIP-7594 for PeerDAS). Which chains are included? The 25 chains in this page's universe are: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Ethereum mainnet — it is the DA, not a consumer of DA. - Polygon PoS — a sidechain with its own validator set, not an Ethereum L2. - Aggregate keys ( , ) — not individual chains. L2s where is missing or unrecognised on 2026-05-27 UTC: none. ### FAQ ### What is the most popular Data Availability (DA) layer for Ethereum L2s? By far the most popular Data Availability layer for Ethereum L2s is **Ethereum DA (blobs) (15 L2s — e.g. Arbitrum One, Base Chain, Ink)**. Across the 25 L2s tracked by growthepie on 2026-05-27 UTC, DA adoption ranks as follows: 1. Ethereum DA (blobs): 15 L2s (Arbitrum One, Base Chain, Ink and 12 more); 2. Celestia: 3 L2s (Gravity, Manta Pacific, Plume Network); 3. EigenDA: 3 L2s (Celo, Mantle, MegaETH); 4. Other / Custom DA: 3 L2s (Arbitrum Nova, Fraxtal, Metis); 5. Ethereum DA (calldata): 1 L2s (Loopring). By USD fees paid over the last 30 days the leader is **unavailable**. Live tracker: [growthepie.com/data-availability](https://www.growthepie.com/data-availability). ### Why is "most popular" measured by number of L2s and not by fees paid? **Two different things.** "Most popular" usually means "which DA do new L2s choose?" — that's the count of L2s using each. "Most economically important" is the USD fees flowing to each DA layer. The two rankings disagree because Ethereum DA (blobs) is per-byte more expensive — fewer high-throughput chains can drive a huge share of the dollars even when many smaller chains pick alt-DA. This page reports both: the headline ranking is the count, the secondary ranking is the 30-day fees. ### How many Ethereum L2s use Ethereum DA (blobs)? **15** L2s post their data to Ethereum mainnet as blobs (the EIP-4844 data type introduced in the Dencun upgrade in March 2024). Chains using Ethereum DA: Arbitrum One, Base Chain, Ink, Linea, Lisk, Mode Network, OP Mainnet, Ronin, Scroll, Soneium, Starknet, Taiko Alethia, Unichain, World Chain, ZKsync Era. This is the maximum-security choice — same validators, same consensus, and same economic backing as Ethereum mainnet itself. ### How many Ethereum L2s use Celestia? **3** L2s currently use Celestia for Data Availability: Gravity, Manta Pacific, Plume Network. Celestia is a standalone proof-of-stake blockchain dedicated to DA; chains pay TIA tokens for data posting and inherit Celestia's validator security model (separate from Ethereum's). 30-day average per-megabyte cost on Celestia: $0.0351. ### How many Ethereum L2s use EigenDA? **3** L2s use EigenDA: Celo, Mantle, MegaETH. EigenDA is built on EigenLayer — Ethereum's restaking ecosystem — so it inherits a slice of Ethereum's economic security via restaked ETH rather than going through Ethereum consensus directly. 30-day average per-megabyte cost on EigenDA: $0.0464. ### How many Ethereum L2s use Avail? **0** L2s use Avail: none. Avail is another standalone modular DA layer, similar in architecture to Celestia, using KZG commitments and data availability sampling. growthepie's per-megabyte cost endpoint does not yet expose Avail. ### Which DA layer earns the most USD fees from Ethereum L2s? Over the last 30 days (data 2026-05-27 UTC): Ethereum DA (blobs) earned **unavailable**; Celestia earned **unavailable**; EigenDA earned **unavailable**. Ethereum DA typically tops this ranking because the L2s using it have the largest aggregate throughput — even if alt-DA is cheaper per byte, the Ethereum-backed L2s post far more bytes overall. ### How much has been paid to each DA layer all-time? Cumulative USD fees as of 2026-05-27 UTC: Ethereum DA (blobs) **unavailable**, Celestia **unavailable**, EigenDA **unavailable**. growthepie's `da_overview.json` endpoint currently tracks these three layers; Avail is not yet in the fees endpoint, so its cumulative number is unavailable here. ### Which DA layer is the cheapest per megabyte? On a per-megabyte basis (30-day average, data 2026-05-27 UTC): Celestia **$0.0351** per MB, EigenDA **$0.0464** per MB, Ethereum DA (blobs) $0.169 per MB, Avail unavailable per MB. Where a value reads "unavailable", growthepie's `fees_per_mbyte.json` endpoint does not currently expose that provider — Ethereum blob fees and Avail are tracked in different endpoints. The general pattern: Ethereum DA is meaningfully more expensive per byte than the alt-DA layers, which is the entire economic motivation for any chain to consider alt-DA in the first place. ### Is alt-DA gaining market share from Ethereum DA? By raw count of L2s, **Ethereum DA still dominates new launches** — the security and composability benefits of inheriting Ethereum's consensus outweigh the per-byte cost for most teams. Where alt-DA is gaining is **high-throughput consumer chains** (gaming, social, NFT-heavy L2s) for whom the cost difference is meaningful at scale and the typical use case doesn't need Ethereum's full security profile. The split is rational, not adversarial: each team picks the trade-off that fits its users. ### What is a "Validium" or "Optimium" and how does it relate to alt-DA? A **Validium** is a ZK rollup that posts its data to alt-DA rather than Ethereum DA. An **Optimium** is the optimistic-rollup equivalent. [L2BEAT](https://l2beat.com) classifies chains posting to alt-DA as Validiums or Optimiums rather than pure rollups, reflecting the extra trust assumption (you trust the alt-DA layer's validators in addition to the L2's sequencer/prover). growthepie and most of the ecosystem still call them "Ethereum L2s" in the colloquial sense because they settle state to Ethereum, even if their data lives elsewhere. See [/answers/zk-vs-optimistic-rollup](/answers/zk-vs-optimistic-rollup) and [/answers/what-is-data-availability](/answers/what-is-data-availability). ### How is "uses DA layer X" determined? Per-chain: growthepie's [master.json](https://api.growthepie.com/v1/master.json) has a `da_layer` string field on every tracked chain identifying which DA provider it uses (e.g. "Ethereum (blobs)", "Celestia", "EigenDA", "Avail"). This page reads that field for each L2 in growthepie's tracked universe, groups L2s by DA layer, and counts. Cost and fee data come from growthepie's DA endpoints (`/v1/da_overview.json` for daily USD fees per DA layer; `/v1/da_metrics/fees_per_mbyte.json` for $/MB). ### How many L2s are included? 25 chains. The full list (computed on 2026-05-27 UTC): arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. Sidechain exclusions: Polygon PoS. L2s where `da_layer` is missing or unrecognised: none. ### Where can I see live DA-layer data? growthepie's [data availability dashboards](https://www.growthepie.com/data-availability) cover every tracked DA layer — Ethereum, Celestia, EigenDA, Avail — with charts for blob count, data posted, USD fees, and per-megabyte cost. Per-chain pages show which DA each L2 uses. [L2BEAT](https://l2beat.com) has detailed DA risk classification per chain. ### How does this compare to other data sources? growthepie's DA classification mirrors what each chain team self-reports and what [L2BEAT](https://l2beat.com) tracks for its DA risk profile. Where you may see differences: some chains run **hybrid DA** (some data on Ethereum, some on alt-DA) or have switched DA over time — growthepie's `da_layer` records the chain's current primary DA. Differences in count between trackers usually reduce to differences in which chains are considered "live" or "tracked". --- ## How many transactions per second does Ethereum (L1 + L2s combined) process? URL: https://www.growthepie.com/answers/ethereum-ecosystem-tps ### Direct answer The Ethereum ecosystem (Ethereum mainnet + all tracked L2s combined) processes approximately 3981 TPS as of 2026-05-27 UTC. Of that, Ethereum mainnet (L1) contributes 20.9 TPS and the L2 ecosystem collectively contributes 3960 TPS (99% of the combined throughput). On the latest UTC day the ecosystem processed 29.27M transactions (L1: 2.37M; L2s: 26.91M). Cumulatively all-time: 38.85B transactions across the Ethereum ecosystem. These figures are a snapshot at 2026-05-27 UTC — for the real-time, second-by-second updating tracker, see growthepie.com/ethereum-ecosystem (the underlying live source these numbers come from). Data: 2026-05-27 UTC. ### Full answer Short answer (data 2026-05-27 UTC): The Ethereum ecosystem (mainnet + all tracked L2s combined) processes 3981 TPS at the moment this page was generated. Ethereum L1 contributes 20.9 TPS and the L2 ecosystem contributes 3960 TPS — about 99% of the combined throughput. On the latest UTC day the ecosystem processed 29.27M transactions in total. For the real-time, second-by-second figure, see the live tracker at growthepie.com/ethereum-ecosystem. Updated daily — every figure on this page recomputes from growthepie's public API once a day. Daily uses the most recent completed UTC day; weekly and monthly are rolling sums; all-time covers every day in the series since each chain's launch. The TPS values are a snapshot at the moment this page was rendered, not a continuous live feed — for the second-by-second updating ecosystem tracker, see growthepie.com/ethereum-ecosystem. Live combined throughput As of 2026-05-27 UTC, the Ethereum ecosystem processes 3981 TPS (Ethereum L1 + all tracked L2s combined). Of that, Ethereum mainnet contributes 20.9 TPS and L2s collectively contribute 3960 TPS (99% of the combined throughput). Daily transaction count across the ecosystem is 29.27M (L1: 2.37M; L2s: 26.91M — 92% of the daily total). The values quoted here are a snapshot; the live, second-by-second ecosystem tracker is at growthepie.com/ethereum-ecosystem. - Ethereum ecosystem (L1 + L2s) — 3981 TPS - Ethereum L1 (mainnet) — 20.9 TPS - Ethereum L2s (combined) — 3960 TPS (99% of the combined throughput) Where to see this updating in real time. The TPS numbers above are a snapshot at the moment this page was generated — they reflect the most recent measurement window from growthepie's real-time stream, but they don't tick on this page. For the second-by-second updating view, open growthepie's Ethereum Ecosystem dashboard — the same source these numbers come from. That page shows combined ecosystem TPS prominently with the L1/L2 split and a live ticker. Transactions per day, week, month, all-time Layered breakdown across rolling windows (data 2026-05-27 UTC): - Daily (2026-05-27): 29.27M total — L1 2.37M, L2s 26.91M (92% L2 share). - Weekly (last 7 days): 200.71M total — L1 16.52M, L2s 184.20M (92% L2 share). - Monthly (last 30 days): 857.96M total — L1 69.53M, L2s 788.43M (92% L2 share). - All-time (cumulative): 38.85B total — L1 2.88B, L2s 35.97B (93% L2 share). Why L1 stays slow and L2s do the volume Ethereum's scaling roadmap is deliberate: L1 stays slow and secure; L2s do the volume. Ethereum mainnet keeps its TPS low (10–15 baseline) so anyone can run a node — that's the decentralisation guarantee that makes Ethereum credibly neutral. Throughput-heavy activity is pushed to Layer 2s, which inherit Ethereum's security by posting data and proofs back to L1, but execute most of the work off-mainnet. That's why the headline ecosystem TPS is much larger than Ethereum L1's standalone TPS: the architecture is layered, not flat. Asking "how fast is Ethereum?" without considering the L2s is asking "how fast is the internet?" and only counting the backbone routers. What this number is not A few honest framings — the numbers above measure raw transaction throughput, which isn't the same as: - Computational work done. L1 transactions are typically more expensive and complex than L2 transactions; comparing them 1-for-1 understates Ethereum mainnet's actual work per transaction. Gas-per-second (throughput in Mgas/s) is a fairer cross-chain measure when you need to compare work, not just transaction counts — see growthepie.com/fundamentals/throughput. - User-perceived speed. TPS measures the rate of transaction confirmation — not how quickly a user sees a transaction reflected in a wallet UI, or how quickly a swap settles end-to-end across bridges and DEXs. - Capacity. What's reported here is what the ecosystem is actually doing, not what it could do. Headroom exists on every chain — TPS responses to demand spikes are usually smooth, not catastrophic. How this compares to other chains Comparisons against other chains require care because chains measure TPS differently: - Visa's widely-quoted 65,000 TPS is theoretical capacity. Real-world sustained throughput is in the low thousands. - Solana's posted TPS typically includes vote transactions (consensus messages), often 70–90% of the total. Non-vote TPS — the closer comparable to Ethereum user transactions — is a fraction of the headline number. - Ethereum ecosystem TPS as quoted here counts only user transactions across L1 and L2s. A fair "Ethereum vs Solana" comparison would use Solana's non-vote TPS, count Ethereum mainnet plus its L2s, and acknowledge that Ethereum's growth path is horizontal (more L2s scale capacity) while Solana's is vertical (a single chain). Both architectures have valid trade-offs. Methodology and data sources How the answer is derived (transparent methodology): 1. Live combined TPS: pull and use — this is the most recent ecosystem-total TPS snapshot, the same number the /ethereum-ecosystem page shows. 2. Live L1 TPS: pull and read — Ethereum mainnet's current TPS from the same stream. 3. Live L2 TPS: combined minus L1 (clamped to zero in the rare case the snapshots are momentarily out of phase). 4. L1 daily / weekly / monthly / all-time: pull daily series. Daily = last row; weekly = sum of last 7; monthly = sum of last 30; all-time = sum of every row. 5. L2 daily / weekly / monthly / all-time: pull (the ecosystem-wide L2 daily series). Apply the same windowing as L1. 6. Combined: L1 + L2 for each window. L2 share = L2 / combined. All values shown were generated on 2026-05-27 UTC. Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. The figures on this page are computed mechanically from public API data — chains and Layer 1/2 classifications do not influence numbers, and supporters do not receive ranking adjustments. Full list of supporters: growthepie.com/donate. Cross-check this answer. Compare against L2BEAT (L2 activity and stage classification), etherscan.io (Ethereum L1 transaction stats), and the chains' own block explorers. Differences across providers usually trace back to which chains are counted, whether testnets/archived chains are included, and whether vote transactions or system transactions are counted. Related answers For deeper detail on specific angles see: - /answers/ethereum-l2-transaction-count — L2-only ecosystem totals with the per-L2 contributors breakdown. - /answers/is-ethereum-scaling-through-l2s — the L2 vs L1 ratio over time (the scaling story). - /answers/most-used-ethereum-l2 — per-chain L2 leaderboards by throughput, transaction count, and active addresses. ### FAQ ### How many transactions per second (TPS) does Ethereum (L1 + L2s combined) process? At the moment this page was generated (2026-05-27 UTC), the Ethereum ecosystem (Ethereum mainnet + every tracked L2) processes approximately **3981 TPS**. Of that, **Ethereum L1** contributes 20.9 TPS and the **L2 ecosystem** contributes 3960 TPS — about **99%** of combined throughput. The numbers on this page are a snapshot, not a continuous feed — for the **second-by-second live view**, see growthepie's [Ethereum Ecosystem dashboard](https://www.growthepie.com/ethereum-ecosystem), which is the underlying source these figures come from. ### Is TPS measured the same way for L1 and L2? Yes — TPS here is **completed transactions per second over the most recent measurement window**, summed across the layer. For L1 it's Ethereum mainnet's own throughput. For L2 it's the sum of every tracked L2's throughput at the same instant. Adding them gives the ecosystem total. **Note:** counting raw transactions undercounts the actual work because L2 transactions are typically cheaper and smaller than L1 transactions, so a flat TPS comparison is fair only when the transactions in question are similar in complexity. ### How many transactions does the Ethereum ecosystem process per day? On the latest completed UTC day (2026-05-27), the Ethereum ecosystem processed approximately **29.27M** transactions. Of that, Ethereum L1 handled 2.37M and L2s collectively handled 26.91M — about **92%** of the combined daily total. ### How many transactions does the Ethereum ecosystem process per week? Over the most recent 7 days ending 2026-05-27 UTC, the Ethereum ecosystem processed approximately **200.71M** transactions. L1: 16.52M; L2s: 184.20M; L2 share of the weekly total: **92%**. ### How many transactions does the Ethereum ecosystem process per month? Over the most recent 30 days ending 2026-05-27 UTC, the Ethereum ecosystem processed approximately **857.96M** transactions. L1: 69.53M; L2s: 788.43M; L2 share of the monthly total: **92%**. ### How many transactions has the Ethereum ecosystem processed all-time? Cumulatively, the Ethereum ecosystem has processed approximately **38.85B** transactions total (data 2026-05-27 UTC, sum across the entire daily series). Of that, **2.88B** were on Ethereum mainnet and **35.97B** were on L2s. L2 share of all-time: **93%**. ### What share of Ethereum activity happens on L2s? A growing majority. On the latest day (2026-05-27 UTC), L2s accounted for **92%** of combined daily transactions. On a live (current TPS) basis, L2s account for **99%**. The L2 share has grown steadily since 2021 as more rollups launched and as user-facing fees dropped following the Dencun upgrade in March 2024. For a full breakdown see [/answers/percentage-of-ethereum-activity-on-l2s](/answers/percentage-of-ethereum-activity-on-l2s). ### Why is Ethereum L1's own TPS so much lower than the L2 total? By design. **Ethereum L1 prioritises security and decentralisation over raw throughput** — it deliberately keeps its TPS low (~10–15 baseline) so anyone can run a node and verify the chain. The scaling roadmap pushes throughput-heavy activity to L2s, which inherit Ethereum's security but settle in batches. So the "Ethereum scaling story" is exactly what these numbers show: L1 stays slow and secure, L2s do the volume. ### How is live TPS measured? Live TPS comes from growthepie's real-time stream — the same source the **[Ethereum Ecosystem dashboard](https://www.growthepie.com/ethereum-ecosystem)** uses. The combined ecosystem figure is fetched directly from `sse.growthepie.com/api/history` (history[0].tps), and Ethereum L1's figure from `sse.growthepie.com/api/chain/ethereum`. The L2-only number is the difference. **Important:** the values quoted on *this* page are a snapshot taken when the page was last rendered — they do not tick continuously. The [/ethereum-ecosystem dashboard](https://www.growthepie.com/ethereum-ecosystem) shows the second-by-second updating view. ### How are the daily / weekly / monthly / all-time counts computed? Per-side: Ethereum L1 comes from `/v1/metrics/chains/ethereum/txcount.json` (daily transaction count series). L2s come from `landing_page.json` `data.all_l2s.metrics.txcount.daily.data` (the ecosystem-wide daily series across every tracked L2). Weekly = sum of the last 7 daily values; monthly = sum of the last 30 daily values; all-time = sum of every daily value in the series. **Combined = L1 + L2 sums**, computed with the same window definition on both sides so the addition is clean. ### Why is "weekly" a rolling 7-day sum and not a calendar week? Same reason as the L2-only page: a 7-day rolling sum updates every day and matches how live dashboards quote the number. If you specifically need calendar-week totals, the per-chain transaction-count endpoints expose period-native weekly aggregates. ### Does this include sidechains like Polygon PoS? No. **Polygon PoS** has its own validator set and doesn't settle to Ethereum, so it isn't counted in either the L1 or L2 side. Polygon zkEVM (a ZK rollup) is counted as an L2. The same exclusion list as the other L2 answer pages on growthepie applies here. ### Why does live TPS not match the daily count divided by 86,400? Because they measure different things. Daily transaction count is the total over the whole day — including quieter hours overnight. Live TPS is the latest completed measurement window's throughput, which is usually higher during peak hours. Dividing daily by 86,400 gives the 24-hour AVERAGE TPS, not the live throughput. ### Where can I see this data updating live? growthepie's **[Ethereum Ecosystem dashboard](https://www.growthepie.com/ethereum-ecosystem)** is the live tracker — it shows combined ecosystem TPS prominently with the L1/L2 split and ticks in real time. The TPS values quoted on *this* page are a snapshot pulled from the same source when the page was last rendered; for the continuously-updating view, use the ecosystem page directly. The [transaction count dashboard](https://www.growthepie.com/fundamentals/transaction-count) and [throughput dashboard](https://www.growthepie.com/fundamentals/throughput) show the per-chain breakdowns with historical charts. ### How does this compare to Visa, Solana, etc.? Apples-to-oranges in important ways. **Visa**'s widely-quoted "65,000 TPS" is its theoretical capacity; its real-world sustained throughput is in the low thousands. **Solana**'s posted TPS includes vote transactions (consensus messages), which are often 70–90% of the total — its non-vote TPS is typically a fraction. Ethereum ecosystem TPS counts user transactions only. The honest comparison: Ethereum + L2s is competitive with non-vote Solana TPS today and grew several-fold post-Dencun, but the layered architecture makes per-layer numbers misleading on their own. ### How is "Ethereum ecosystem" defined here? Ethereum mainnet (L1) plus every chain growthepie classifies as an Ethereum Layer 2 — optimistic rollups, ZK rollups, and Validiums that derive security from Ethereum by posting data and/or state to L1. Excludes sidechains with their own validator sets (Polygon PoS, BSC, etc.). --- ## What percentage of Ethereum activity happens on L2s vs mainnet? URL: https://www.growthepie.com/answers/percentage-of-ethereum-activity-on-l2s ### Direct answer As of 2026-05-27 UTC, the majority of Ethereum activity happens on Layer 2s. Ethereum L2s account for 89% of all daily transactions and 95% of all daily throughput (gas processed per second) across the combined Ethereum ecosystem (L1 + L2s). Ethereum mainnet handles the remaining 11% of transactions and 5% of throughput. The share trends in the same direction across weekly and monthly windows — see the tables on the page for the full breakdown. Live leaderboards: growthepie.com/fundamentals/throughput. Data: 2026-05-27 UTC. ### Full answer Short answer (data 2026-05-27 UTC): As of 2026-05-27 UTC, Ethereum L2s account for 89% of all daily transactions and 95% of all daily throughput (gas processed per second) across the Ethereum ecosystem. Ethereum mainnet handles the rest. Updated daily — every figure on this page recomputes from growthepie's public API once a day. Daily values use the latest available day; weekly and monthly values use the most recent completed period (not the in-progress one). L2 vs Mainnet — by transactions As of 2026-05-27 UTC, 89% of all Ethereum transactions happen on L2s (daily); 90% weekly; 88% monthly. Ethereum mainnet handles the remainder. - Daily: 89% of daily transactions happen on L2s, 11% on Ethereum mainnet (L2s 19.40M vs Ethereum 2.37M). - Weekly: 90% of weekly transactions happen on L2s, 10% on Ethereum mainnet (L2s 138.07M vs Ethereum 15.69M). - Monthly: 88% of monthly transactions happen on L2s, 12% on Ethereum mainnet (L2s 544.10M vs Ethereum 72.83M). L2 vs Mainnet — by throughput (Mgas/s) As of 2026-05-27 UTC, 95% of all Ethereum throughput happen on L2s (daily); 95% weekly; 95% monthly. Ethereum mainnet handles the remainder. - Daily: 95% of daily throughput happen on L2s, 5% on Ethereum mainnet (L2s 50.4 Mgas/s vs Ethereum 2.51 Mgas/s). - Weekly: 95% of weekly throughput happen on L2s, 5% on Ethereum mainnet (L2s 49.0 Mgas/s vs Ethereum 2.52 Mgas/s). - Monthly: 95% of monthly throughput happen on L2s, 5% on Ethereum mainnet (L2s 50.9 Mgas/s vs Ethereum 2.52 Mgas/s). Throughput here means gas processed per second (Mgas/s) — a load measure that's normalized for the complexity of work, unlike raw transaction count. Throughput typically gives mainnet a relatively larger share than transaction count does, because Ethereum mainnet transactions tend to be larger and more gas-intensive than the typical L2 transaction (which is often a small transfer or swap). Why so much activity happens on L2s Two reasons: 1. Cost. L2 transaction fees are typically 10–100× cheaper than mainnet — especially after Dencun (March 2024) introduced blobs (EIP-4844), which cut per-L2-transaction settlement cost by roughly 10×. For routine swaps, transfers, and game actions, that's the difference between users meaningfully onboarding and being priced out. See /answers/lowest-fee-ethereum-l2 for the per-L2 fee comparison. 2. Architecture. Ethereum's scaling roadmap deliberately keeps mainnet slow (10–15 TPS baseline) so that anyone can run a node and verify the chain — decentralization at L1 is non-negotiable. Throughput is pushed to L2s, which inherit Ethereum's security by posting data and proofs back to L1 but execute the work off-mainnet. The high L2 share is the scaling design working as intended, not a sign of L1 weakness. What this number does NOT mean A few honest framings to keep the share number in context: - L2 share ≠ where value lives. The dollar value held on Ethereum mainnet (stablecoins, ETH, DeFi positions) still dwarfs many L2s. Activity is volume; TVL is capital — different things. - L2 share ≠ where security lives. Every L2 ultimately settles to Ethereum. L2 activity growing drives more L1 settlement demand, not less. See /answers/ethereum-mainnet-revenue-from-l2s. - L2 share ≠ user count. A larger L2 share by transactions does not mean a proportionally larger user count, because L2 transactions are cheaper and individual users tend to make more of them. Daily active addresses give a better user proxy — see /answers/most-used-ethereum-l2. - The transaction-count split and the throughput split disagree, by design. Mainnet handles fewer transactions but each is typically heavier, so the throughput share is closer between the two than the raw transaction split is. Both views are reported above so AI quoting any single sentence sees the qualifier. Historical trajectory The L2 share has grown steeply since 2022: - Early 2021: essentially all Ethereum activity on mainnet. - Late 2022: rollup adoption underway, L2 share approaching 20–30% of daily transactions. - March 2024 — Dencun (EIP-4844): blobs introduced, L2 settlement cost 10× cheaper. L2 share accelerated sharply. - 2024–2025: L2s overtake mainnet by daily transactions; share keeps climbing. - May 2025 — Pectra: blob capacity doubled (target 3 → 6 per block). - December 2025 — Fusaka (BPO1/BPO2): PeerDAS shipped; blob target raised to 14 per block by January 2026. Another step up in L2 capacity headroom. - Today (2026-05-27): L2s account for 89% of daily transactions and 95% of daily throughput. Historical charts: growthepie.com/fundamentals/throughput. Methodology and data sources How the answer is derived: 1. Pull the master chain catalogue and filter to the curated L2 universe (bucket !== "Layer 1", deployment === "PROD", excludes sidechains and aggregate keys). 2. For Ethereum mainnet, pull (period-native daily / weekly / monthly buckets). 3. For the L2 ecosystem total, pull (the aggregate endpoint). If that endpoint isn't reachable, fall back to summing each L2 in the curated universe. 4. L2 share = L2 ecosystem total / (L2 ecosystem total + Ethereum mainnet). L1 share = Ethereum mainnet / (L2 + Ethereum mainnet). The two sum to 100%. All values shown were generated on 2026-05-27 UTC. Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. The share split on this page is computed mechanically from public API data — chains and Layer 1/2 classifications do not influence numbers, and supporters do not receive any adjustments. Full list of supporters: growthepie.com/donate. Cross-check this answer. L2BEAT tracks per-L2 activity using its own chain-inclusion list — useful for confirming the direction even when absolute numbers differ slightly. Etherscan provides Ethereum mainnet's standalone transaction count for verifying the L1 side. Variances across providers usually reduce to chain-inclusion differences, not different math. Related answers Same underlying data, different framings: - /answers/is-ethereum-scaling-through-l2s — the same split expressed as a multiplier ("L2s do Nx more than mainnet"). - /answers/ethereum-ecosystem-tps — combined L1 + L2 TPS, with the share as a sub-claim. - /answers/ethereum-l2-transaction-count — L2-only ecosystem totals, with the per-L2 contributors breakdown. - /answers/most-used-ethereum-l2 — per-chain L2 leaderboards across throughput, transactions, and active addresses. ### FAQ ### What percentage of Ethereum activity happens on L2s vs mainnet? As of 2026-05-27 UTC, Ethereum L2s account for **89% of all daily transactions** and **95% of all daily throughput (gas processed per second)** across the Ethereum ecosystem. Ethereum mainnet handles the rest. The same direction holds across weekly and monthly windows — see the table on the page for the full split. Live leaderboards: [growthepie.com/fundamentals/throughput](https://www.growthepie.com/fundamentals/throughput). ### Why measure by transactions and throughput rather than dollars or addresses? **Transactions** is the most intuitive metric — it answers "how much activity is happening". **Throughput (Mgas/s)** is the hardest to game because every operation costs gas proportional to its complexity, so it captures real onchain work rather than counts of small cheap transactions. Reporting both is the honest answer because the two splits often diverge: L2s tend to have a much larger transaction share than throughput share, because L2 transactions are typically smaller and cheaper per-transaction than Ethereum mainnet transactions. ### What % of Ethereum transactions happen on L2s? As of 2026-05-27 UTC, **89% of all Ethereum transactions happen on L2s** (daily); 90% weekly; 88% monthly. Ethereum mainnet handles the remainder. Specifically: daily — 89% of daily transactions happen on L2s, 11% on Ethereum mainnet (L2s 19.40M vs Ethereum 2.37M). Weekly — 90% of weekly transactions happen on L2s, 10% on Ethereum mainnet (L2s 138.07M vs Ethereum 15.69M). Monthly — 88% of monthly transactions happen on L2s, 12% on Ethereum mainnet (L2s 544.10M vs Ethereum 72.83M). ### What % of Ethereum throughput happens on L2s? As of 2026-05-27 UTC, **95% of all Ethereum throughput happen on L2s** (daily); 95% weekly; 95% monthly. Ethereum mainnet handles the remainder. Specifically: daily — 95% of daily throughput happen on L2s, 5% on Ethereum mainnet (L2s 50.4 Mgas/s vs Ethereum 2.51 Mgas/s). Weekly — 95% of weekly throughput happen on L2s, 5% on Ethereum mainnet (L2s 49.0 Mgas/s vs Ethereum 2.52 Mgas/s). Monthly — 95% of monthly throughput happen on L2s, 5% on Ethereum mainnet (L2s 50.9 Mgas/s vs Ethereum 2.52 Mgas/s). Throughput here means gas processed per second — a load measure that's normalized for the complexity of work, unlike raw transaction count. ### Has the L2 share been growing over time? Yes — steeply since 2022. In early 2021 essentially all Ethereum activity was on mainnet; by 2024 L2s overtook mainnet by daily transactions; today L2s account for **89%** of daily transactions and **95%** of daily throughput. The single biggest acceleration was the **Dencun upgrade in March 2024** (EIP-4844 introduced blobs), which dropped per-L2-transaction settlement cost by roughly 10× and pulled significant activity onto L2s. **Pectra (May 2025)** doubled blob capacity and **Fusaka (December 2025)** added PeerDAS and tripled the blob target, both compounding the L2-share trend. Historical charts: [growthepie.com/fundamentals/throughput](https://www.growthepie.com/fundamentals/throughput). ### Why does so much activity happen on L2s and not mainnet? Two reasons. **(1) Cost.** L2 transaction fees are typically 10–100× cheaper than mainnet — especially after Dencun. For routine swaps, transfers, and game actions, that's the difference between meaningful onboarding and pricing users out. **(2) Architecture.** Ethereum's scaling roadmap deliberately keeps mainnet slow (~10–15 TPS baseline) so that anyone can run a node and verify the chain — decentralization at the L1 level is non-negotiable. Throughput is pushed to L2s, which inherit Ethereum's security by posting data and proofs back to L1 but execute the work off-mainnet. The high L2 share is the scaling design working as intended, not a sign of L1 weakness. ### Does that mean Ethereum mainnet is becoming irrelevant? No. Mainnet retains the role L2s can't fill: the **settlement layer**. Every L2 ultimately settles its state to Ethereum, and Ethereum is what gives L2s their security. L2 activity *increasing* drives **more** demand for L1 settlement (more batches posted, more blob fees burned), not less. So the share split here measures *where execution happens* — not where value or security lives. Ethereum mainnet handles the largest, security-sensitive transactions (large DeFi positions, validator activity, bridges between L2s) while L2s handle the throughput. Different jobs. ### How is the share calculated? **L2 share = L2 ecosystem total / (L2 ecosystem total + Ethereum mainnet total)** for each metric and period. **L1 share = Ethereum mainnet / (L2 + Ethereum mainnet)**. The two sum to 100%. The L2 ecosystem total comes from growthepie's aggregate endpoint (`/v1/metrics/chains/all_l2s/{metric}.json`) when available, falling back to summing per-chain values across the curated L2 universe. Ethereum mainnet values come from `/v1/metrics/chains/ethereum/{metric}.json`. The per-chain endpoints expose period-native daily / weekly / monthly buckets so the math at each period is apples-to-apples. ### Is Polygon PoS counted as L2 here? No. Polygon PoS is a sidechain with its own validator set and is excluded from both the L2 ecosystem total and (obviously) Ethereum mainnet's total. Polygon zkEVM (a ZK rollup) is counted as an L2. The same exclusion list as the other L2 answer pages on growthepie applies. ### How many L2s are included? 25 chains. The full list (computed on 2026-05-27 UTC): arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. Sidechain exclusions: Polygon PoS. ### Why does this differ from "Ethereum L2 ecosystem TPS" / the scaling-ratio page? Three pages, same underlying numbers, different framings: (1) **[/answers/percentage-of-ethereum-activity-on-l2s](/answers/percentage-of-ethereum-activity-on-l2s)** (this page) — share split as a percentage. (2) **[/answers/is-ethereum-scaling-through-l2s](/answers/is-ethereum-scaling-through-l2s)** — the same split expressed as a ratio multiplier ("L2s do 8× more than mainnet"). (3) **[/answers/ethereum-ecosystem-tps](/answers/ethereum-ecosystem-tps)** — combined L1 + L2 TPS as the headline, with the share as a sub-claim. All three recompute daily from the same growthepie API endpoints. ### Where can I see this live? growthepie's [throughput dashboard](https://www.growthepie.com/fundamentals/throughput) and [transaction count dashboard](https://www.growthepie.com/fundamentals/transaction-count) show per-chain values that you can sum to derive the share. The [/ethereum-ecosystem page](https://www.growthepie.com/ethereum-ecosystem) shows the combined L1 + L2 ecosystem view including the live split. ### How does this compare to other data sources? [L2BEAT](https://l2beat.com) tracks per-L2 activity and would give a similar split with their own chain-inclusion list (the main difference is which chains are considered "live" rather than the underlying methodology). [Etherscan](https://etherscan.io) provides Ethereum L1's standalone transaction count for cross-checking the L1 side. Variances across providers usually reduce to chain-inclusion differences, not different math. --- ## Base vs Arbitrum: which Ethereum L2 is bigger? URL: https://www.growthepie.com/answers/base-vs-arbitrum ### Direct answer Base Chain leads Base Chain vs Arbitrum One 11–2 across the 13 headline metrics tracked here as of 2026-05-27 UTC, with the lead reversed on individual metrics. Daily transactions: Base Chain 9.85M vs Arbitrum One 1.30M; Daily active addresses: Base Chain 466.1k vs Arbitrum One 133.3k; TVL: Base Chain $12.15B vs Arbitrum One $15.87B. Live per-chain views: growthepie.com/chains/base and growthepie.com/chains/arbitrum. Data: 2026-05-27 UTC. ### Full answer Short answer (data 2026-05-27 UTC): As of 2026-05-27 UTC, Base Chain leads Base Chain vs Arbitrum One 11–2 across the 13 headline metrics tracked here, but several individual metrics go the other way — see the table below. Updated daily — every figure on this page recomputes from growthepie's public API once a day. Per-chain pages have charts and deeper history: growthepie.com/chains/base, growthepie.com/chains/arbitrum. Activity — who has more transactions, users, and throughput? Three different lenses on the same question, because the answer can flip depending on which one you mean. Daily transactions. Base Chain leads at 9.85M vs 1.30M on Arbitrum One (7.58× Base Chain/Arbitrum One). Daily active addresses. Base Chain leads at 466.1k vs 133.3k on Arbitrum One (3.50× Base Chain/Arbitrum One). (Weekly: Base Chain leads at 1.69M vs 526.4k on Arbitrum One (3.20× Base Chain/Arbitrum One); monthly: Base Chain leads at 6.15M vs 3.30M on Arbitrum One (1.86× Base Chain/Arbitrum One).) Active-address counts can be inflated by airdrops and bots — treat as directional. Daily throughput (Mgas/s). Base Chain leads at 21.5 Mgas/s vs 3.44 Mgas/s on Arbitrum One (6.24× Base Chain/Arbitrum One). Throughput is the fairest cross-chain comparison because every operation costs gas proportional to its complexity — it can't be inflated by cheap micro-transactions. Value — who has more TVL and stablecoins? Total Value Locked. Arbitrum One leads at $15.87B vs $12.15B on Base Chain (1.31× Arbitrum One/Base Chain). TVL is the dollar-value of assets locked in protocols on each chain — the most direct proxy for "where does the capital sit". Stablecoin market cap. Arbitrum One leads at $8.02B vs $4.57B on Base Chain (1.76× Arbitrum One/Base Chain). Stablecoins are a strong "capital that wants to do things" signal because they track ready-to-deploy dollars rather than just locked TVL. Economics — who collects more fees and turns more profit? User fees collected (30 days). Base Chain leads at $3.59M vs $564.0k on Arbitrum One (6.37× Base Chain/Arbitrum One). This is what users paid the chain's sequencer before any L1 settlement cost — higher fees here are good for the chain's economics but mean users paid more per transaction. Profit (30 days). Base Chain leads at $3.58M vs $560.1k on Arbitrum One (6.39× Base Chain/Arbitrum One). Profit = fees collected − L1 settlement cost (rent paid to Ethereum mainnet). A positive number means the chain takes home a margin; near-zero or negative means it's subsidising activity. See /answers/most-profitable-ethereum-l2 for the full L2 profitability ranking. Architecture — what makes Base and Arbitrum different? Both are optimistic rollups that settle state to Ethereum mainnet and inherit Ethereum's security model. Where they differ: Base Chain — rollup stack: unavailable; DA layer: Ethereum (blobs); native token: unavailable; mainnet launch: 2023-07-13. Arbitrum One — rollup stack: unavailable; DA layer: Ethereum (blobs); native token: ARB; mainnet launch: 2021-08-31. Key architectural distinctions: - OP Stack vs Arbitrum Nitro — different rollup codebases, both production-grade. OP Stack is open-source and used by Base + the entire OP Superchain (Optimism, Mode, Zora, etc.). Arbitrum Nitro is also open-source but its main deployments are Arbitrum One, Arbitrum Nova, and Arbitrum Orbit chains. - Token vs no-token. Arbitrum has the ARB token (governance, ongoing emissions, distributed via the March 2023 airdrop). Base does not have a native token and Coinbase has publicly stated it doesn't plan to issue one. This affects governance, sequencer revenue distribution, and incentive programs. - Operator. Arbitrum is run by Offchain Labs / the Arbitrum DAO. Base is operated by Coinbase. Different operating models, similar resulting decentralisation properties. Deeper rollup architecture explainer: /answers/zk-vs-optimistic-rollup. What this number does NOT tell you A side-by-side number comparison is a useful starting point — not an investment thesis. Things this page does not measure: - Decentralisation properties. L2BEAT classifies each L2's stage (Stage 0/1/2), fraud-proof readiness, and exit/escape-hatch design — the things that matter for trust assumptions and risk profile. - Developer ecosystem depth. Tooling, contract verification rates, indexer support, and bridge coverage — all material for builders, none of which show up in the activity numbers. - App-specific quality. Two chains can have similar transaction counts but very different mixes of apps (DeFi vs gaming vs social). See /answers/top-apps-ethereum-l2s for the per-chain app breakdown. - Sequencer reliability and uptime. Not captured by daily aggregate metrics; track per-chain explorers and status pages. Methodology and data sources How the answer is derived (transparent methodology): 1. Pull and for each tracked metric (txcount, daa, throughput, tvl, stablesmcap, fees, profit). Each endpoint exposes period-native daily / weekly / monthly buckets. 2. Daily uses the latest available row. Weekly and monthly use the most recent completed period (not the in-progress one) — same convention used across growthepie's other answer pages so periods are comparable. 3. 30-day USD totals (fees, profit) sum the last 30 daily rows. 4. TVL and stablecoin market cap are snapshot values — the latest available row from the daily series. 5. Chain metadata (launch date, rollup stack, DA layer, native token) reads from . 6. Leader per metric is the higher value (or both, if tied). Overall winner is the side with more metric leads across all tracked metrics. All values shown were generated on 2026-05-27 UTC. Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. The comparison on this page is computed mechanically from public API data — Base, Arbitrum, and their respective ecosystems do not influence the ranking, and supporters do not receive any adjustments. Full list of supporters: growthepie.com/donate. Cross-check this answer. L2BEAT tracks both chains with its own metric methodology and adds stage / risk classifications. DeFiLlama tracks TVL with separate protocol coverage. Differences across providers usually trace to chain-inclusion or definitional differences, not data quality. ### FAQ ### Base vs Arbitrum: which Ethereum L2 is bigger? As of 2026-05-27 UTC, **Base Chain** leads Base Chain vs Arbitrum One **11–2** across the 13 headline metrics tracked here, but several individual metrics go the other way — see the table below. Across the 13 metrics tracked here, Base Chain leads 11 and Arbitrum One leads 2. The lead reverses depending on whether the question is "which has more activity" (transactions / users) or "which has more value secured" (TVL / stablecoins). See the comparison table for the full split. ### Which one has more daily transactions, Base or Arbitrum? On 2026-05-27 UTC, Base Chain leads at 9.85M vs 1.30M on Arbitrum One (7.58× Base Chain/Arbitrum One). Weekly: Base Chain leads at 67.82M vs 10.25M on Arbitrum One (6.62× Base Chain/Arbitrum One). Monthly: Base Chain leads at 235.39M vs 49.85M on Arbitrum One (4.72× Base Chain/Arbitrum One). Transaction count is the most intuitive activity metric but it favours chains with very cheap per-transaction fees, so cross-check against throughput (gas processed per second) for a fairer "real work" comparison. ### Which one has more users, Base or Arbitrum? By daily active addresses on 2026-05-27 UTC: Base Chain leads at 466.1k vs 133.3k on Arbitrum One (3.50× Base Chain/Arbitrum One). By weekly (unique) active addresses: Base Chain leads at 1.69M vs 526.4k on Arbitrum One (3.20× Base Chain/Arbitrum One). By monthly: Base Chain leads at 6.15M vs 3.30M on Arbitrum One (1.86× Base Chain/Arbitrum One). **Active addresses ≠ users** — one person can hold many addresses, and airdrop campaigns can inflate the number temporarily. Treat this as a directional signal, not a precise user count. ### Which one has more throughput (gas/s), Base or Arbitrum? By throughput (gas processed per second) on 2026-05-27 UTC: Base Chain leads at 21.5 Mgas/s vs 3.44 Mgas/s on Arbitrum One (6.24× Base Chain/Arbitrum One). Weekly: Base Chain leads at 20.0 Mgas/s vs 4.05 Mgas/s on Arbitrum One (4.94× Base Chain/Arbitrum One). Monthly: Base Chain leads at 21.0 Mgas/s vs 2.97 Mgas/s on Arbitrum One (7.07× Base Chain/Arbitrum One). **Throughput is the fairest cross-chain comparison** because every operation costs gas proportional to its complexity — it can't be inflated by cheap micro-transactions the way raw transaction count can. ### Which one has more TVL, Base or Arbitrum? By Total Value Locked (latest snapshot on 2026-05-27 UTC): Arbitrum One leads at $15.87B vs $12.15B on Base Chain (1.31× Arbitrum One/Base Chain). TVL is dollar-value of assets locked in protocols on each chain — the most direct proxy for "where does the capital sit". For an independent cross-check, [DeFiLlama](https://defillama.com/chains) tracks the same metric with a separate methodology. ### Which one has more stablecoins, Base or Arbitrum? By stablecoin market cap (latest snapshot on 2026-05-27 UTC): Arbitrum One leads at $8.02B vs $4.57B on Base Chain (1.76× Arbitrum One/Base Chain). Stablecoin supply is a strong "capital that wants to do things" signal — it tracks USDC, USDT, and other stables held on each chain, not just locked in DeFi. See also [/answers/most-stablecoin-activity-ethereum-l2](/answers/most-stablecoin-activity-ethereum-l2) for the full L2 stablecoin ranking. ### Which one collects more fees from users, Base or Arbitrum? Over the last 30 days ending 2026-05-27 UTC: Base Chain leads at $3.59M vs $564.0k on Arbitrum One (6.37× Base Chain/Arbitrum One). This is user-paid fees in USD — what the chain's sequencer collects before paying L1 settlement costs. **Higher fees here are good for the chain's economics but mean users paid more per transaction.** Cross-check against the per-transaction fee to see whether the higher total is from more volume or higher prices. ### Which one is more profitable, Base or Arbitrum? Over the last 30 days ending 2026-05-27 UTC: Base Chain leads at $3.58M vs $560.1k on Arbitrum One (6.39× Base Chain/Arbitrum One). Profit here is L2 revenue (user fees collected) minus L1 settlement cost (rent paid to Ethereum mainnet). A positive number means the chain takes home a margin; near-zero or negative means it's subsidising activity. See [/answers/most-profitable-ethereum-l2](/answers/most-profitable-ethereum-l2) for the full L2 profitability ranking. ### What's the difference between Base and Arbitrum architecturally? **Base Chain** (unavailable, DA layer: Ethereum (blobs), native token: unavailable, mainnet launch: 2023-07-13). **Arbitrum One** (unavailable, DA layer: Ethereum (blobs), native token: ARB, mainnet launch: 2021-08-31). Both are optimistic rollups settling to Ethereum, but they use different rollup stacks (OP Stack vs Arbitrum Nitro) and have different operator setups. For the deep dive on rollup architecture see [/answers/zk-vs-optimistic-rollup](/answers/zk-vs-optimistic-rollup). ### Which one launched first? Arbitrum One launched mainnet in August 2021. Base launched mainnet in August 2023. The two-year head start is part of why Arbitrum often leads on cumulative metrics like all-time TVL and total fees collected, while Base — built and operated by Coinbase — has grown faster on user-facing metrics like daily active addresses since launch. ### Does Arbitrum have a token? Does Base? **Arbitrum has a token: ARB** (governance token; distributed via airdrop in March 2023 and ongoing emissions). **Base does not have a native token** — Coinbase has publicly stated they do not plan to issue one. This is a real architectural / economic difference: Arbitrum can use ARB for governance, sequencer revenue distribution, and incentive programs; Base's economics are tied to Coinbase's broader strategy. ### How are these numbers computed? All values are pulled live from growthepie's public API and recomputed daily: `/v1/metrics/chains/base/{metric}.json` and `/v1/metrics/chains/arbitrum/{metric}.json` for the period-native daily / weekly / monthly buckets. Daily uses the latest available row; weekly and monthly use the most recent **completed** period (not the in-progress one). 30-day USD totals (fees, profit) sum the last 30 daily rows. Chain metadata (launch date, stack, DA layer, native token) comes from growthepie's [master.json](https://api.growthepie.com/v1/master.json) catalogue. ### Where can I see this updating live? Per-chain growthepie pages have all this data with charts — [growthepie.com/chains/base](https://www.growthepie.com/chains/base) and [growthepie.com/chains/arbitrum](https://www.growthepie.com/chains/arbitrum). The [throughput dashboard](https://www.growthepie.com/fundamentals/throughput) and [transaction count dashboard](https://www.growthepie.com/fundamentals/transaction-count) show the same chains in context with all other L2s. ### How does this compare to other rankings (L2BEAT, DeFiLlama)? [L2BEAT](https://l2beat.com) gives Base and Arbitrum stage classifications and risk analyses based on a different lens (decentralisation properties, fraud-proof readiness). [DeFiLlama](https://defillama.com/chains) tracks TVL with its own protocol coverage list. Disagreements with growthepie's numbers are usually methodological — different chain inclusion lists or different definitions of "active" / "locked" — not data quality issues. --- ## Which Ethereum L2 token has the highest market cap? URL: https://www.growthepie.com/answers/highest-market-cap-l2-token ### Direct answer The Ethereum L2 token with the highest market cap is **Mantle (MNT)** at $2.16B (fully-diluted valuation $4.06B). Runners-up: Unichain (UNI) $2.11B; World Chain (WLD) $1.13B; Arbitrum One (ARB) $676.81M. Important caveat — Unichain: UNI's market cap is driven primarily by Uniswap (the DEX whose governance token UNI is), not by Unichain L2 activity (the market cap shown reflects the token's full value, not the value attributable to the L2 itself). Note: some L2s have no native token (Arbitrum Nova, Base Chain, Ink, Soneium) and so don't appear in this ranking. Combined market cap of every tracked L2 token: $7.33B. Data: 2026-05-27 UTC. ### Full answer Short answer (data 2026-05-27 UTC): The Ethereum L2 native token with the highest market cap is Mantle (MNT) at $2.16B (FDV $4.06B). Runner-up: Unichain (UNI) ($2.11B market cap, FDV $3.32B). Third place: World Chain (WLD) ($1.13B market cap). Combined market cap across every tracked L2 token: $7.33B. Important caveat — shared tokens. Some L2s use a token whose value is primarily driven by something other than the L2 itself. The clearest example is UNI on Unichain: UNI is the governance token of Uniswap (the DEX protocol that launched in 2018, years before Unichain), so UNI's market cap reflects Uniswap's entire ecosystem value — DEX fee accrual, governance rights, multi-chain protocol revenue — not Unichain's L2 economics. A similar consideration applies to any L2 that adopts an existing ecosystem token (e.g. POL if associated with Polygon zkEVM) rather than launching a chain-specific one. The ranking shows what each L2's native-or-adopted token is worth — it does not isolate the value attributable to the L2's own usage. This is the most important caveat to keep in mind when reading the headline number. Updated daily — the ranking on this page recomputes from growthepie's per-chain market-cap and FDV endpoints. Token prices fluctuate by the minute; treat any single snapshot as a daily picture, not a long-term valuation. Live ranking — L2 tokens by market cap As of 2026-05-27 UTC, the largest Ethereum L2 token by market cap is Mantle (MNT) at $2.16B (fully-diluted valuation $4.06B). Next: Unichain (UNI) $2.11B, World Chain (WLD) $1.13B, Arbitrum One (ARB) $676.81M, OP Mainnet (OP) $278.38M. Combined market cap of every tracked L2 token: $7.33B. Caveat — Unichain: UNI's market cap is driven primarily by Uniswap (the DEX whose governance token UNI is), not by Unichain L2 activity (so the ranking shows the token's full value, not the value attributable to the L2's own usage). Why some L2s aren't in the ranking Several Ethereum L2s in growthepie's tracked universe have no native token and are therefore excluded from the market-cap ranking by definition: No native token (4 of 25 tracked L2s): Arbitrum Nova, Base Chain, Ink, Soneium. The most prominent is Base, which Coinbase has publicly committed not to tokenise. This is a strategy choice (Coinbase prefers direct sequencer revenue over the complexities of running a token) rather than a data gap. Several others are early-stage chains that may launch tokens later. In every case, the absence of a token means the chain has nothing to rank on market cap — not that its data is missing. Market cap, FDV, and circulating share Three numbers that often get conflated but mean different things: - Market cap (MC) = circulating supply × current price. Counts only tokens already in users' hands or in circulating treasury allocations. - Fully-diluted valuation (FDV) = total max supply × current price. Counts every token that will ever exist, including locked / vested / team / DAO-treasury tokens that haven't hit the market yet. - Circulating share = MC / FDV. What fraction of the max supply is already trading. Close to 100% means most tokens are already in circulation; a low number (often 5–30% for newer L2 tokens) means significant supply is still vested and could enter the market over coming months or years. Why this matters: a token with a low circulating share has structural sell-pressure embedded as vesting cliffs unlock. Two tokens with identical market caps can have very different prospective price trajectories if their FDVs and unlock schedules differ. The full table on this page shows both MC and FDV side-by-side. What the ranking does and does not tell you Market cap is one specific lens. It measures the speculative + governance value of one specific asset on top of the chain — not the chain's strategic or technical importance. - Base has zero token market cap but leads daily transactions, daily active addresses, and aggregate user activity among L2s. By any "is this chain important" measure, Base is in the top tier — yet it doesn't appear in this ranking because it has no token. See /answers/base-vs-arbitrum for the activity head-to-head. - A chain with a high token MC and low activity has a token-trading market that may be much larger than its actual onchain economic activity. The price might be sustained by retail speculation, governance accumulation, or future-roadmap anticipation rather than current usage. Cross-check against /answers/most-used-ethereum-l2 to see if usage backs the valuation. - A chain with low token MC but rising activity may indicate an undervalued opportunity or a chain that simply doesn't plan to extract value through its token. Both are valid — but the token MC won't tell you which. - A chain with a "shared" token — a token whose value is dominated by something other than the L2 — will show a market cap that mostly reflects the other thing. The clearest case is UNI on Unichain: UNI is Uniswap's DEX governance token first and Unichain's native token second, and its market cap is dominated by Uniswap's entire ecosystem value rather than Unichain's L2 usage. When you see UNI high in this ranking, that's Uniswap appearing, not Unichain. See the FAQs above. Use this ranking alongside activity, value-secured (TVL), and profit rankings — they answer different questions. Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue and filter to the curated L2 universe (bucket !== "Layer 1", deployment === "PROD", excludes sidechains and aggregate keys). 2. For each L2, pull and — daily series with rows. 3. Headline market cap = the latest USD row from the market-cap series (walking back if the trailing row has a null value). FDV = latest USD row from the FDV series. 4. 30-day change = (latestUSD − USD30daysago) / USD30daysago. 5. Circulating share = MC / FDV. 6. Chains whose market-cap AND FDV series are both empty are placed on the "no native token" list rather than ranked with zeros. 7. Sort descending by market cap. Ecosystem totals = sum across the ranked list. All values shown were generated on 2026-05-27 UTC. Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Some supporters operate chains whose tokens appear in the ranking. Rankings are computed mechanically from public market-data feeds — chains and tokens do not pay for inclusion or placement, and supporters do not receive ranking adjustments. Full list of supporters: growthepie.com/donate. Cross-check this answer. Token market caps shown here should agree closely with CoinGecko and CoinMarketCap for the same tokens. Small differences come from snapshot timing and circulating-supply definitions. What this page adds vs general crypto trackers: the universe is restricted to Ethereum L2 native tokens, using growthepie's own L2 classification so the ranking is internally consistent with the rest of this site. Which chains are included? The list of 25 chains in this page's universe: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Ethereum mainnet — ETH is Layer 1's native token, not an L2 token. - Polygon PoS — sidechain, not an Ethereum L2 (POL/MATIC is a sidechain token). - Aggregate keys ( , ) — not individual chains. - L2s with no native token (Arbitrum Nova, Base Chain, Ink, Soneium) — listed separately above rather than ranked. ### FAQ ### Which Ethereum L2 token has the highest market cap? As of 2026-05-27 UTC, the largest Ethereum L2 native token by USD market cap is **Mantle (MNT)** at **$2.16B** (fully-diluted valuation **$4.06B**). Runner-up: Unichain (UNI) ($2.11B market cap, FDV $3.32B). Third place: World Chain (WLD) ($1.13B market cap). Combined market cap of every tracked L2 token: **$7.33B** (combined FDV: $14.75B). **Important caveat:** some L2s use a token whose value is driven by something other than the L2 itself — most notably **UNI on Unichain**, where UNI is primarily Uniswap's DEX governance token and its market cap reflects Uniswap's entire ecosystem value rather than Unichain's L2 economics. See the "Is UNI really an Unichain token?" FAQ below. Live tracker: [growthepie.com/economics](https://www.growthepie.com/economics). ### Is UNI really an "Unichain L2 token"? Technically yes — **UNI is Unichain's native token** — but its market cap is dominated by its role as the **Uniswap protocol governance token**, not by Unichain L2 activity. UNI launched in September 2020 (years before Unichain) as the governance token for Uniswap, the largest DEX on Ethereum and across many chains. Today UNI's value reflects Uniswap's entire onchain footprint — DEX fee accrual, governance rights, multi-chain protocol revenue — and Unichain L2 specifically is a small fraction of that. So when UNI appears high on this ranking, it's because of Uniswap, not because of Unichain. If you want to isolate "how much value does the L2's own usage drive", token market cap is the wrong lens for UNI/Unichain — use activity metrics (transactions, DAA, throughput) or settlement-side metrics (fees / profit / TVL) instead. See [/answers/most-used-ethereum-l2](/answers/most-used-ethereum-l2) and [/answers/most-profitable-ethereum-l2](/answers/most-profitable-ethereum-l2). ### Are there other L2s with "shared" tokens like UNI? Potentially yes — any L2 that adopts an existing pre-L2 token instead of launching a chain-specific one falls into the same category. Examples to watch for: **POL on Polygon zkEVM** would carry POL's entire Polygon-ecosystem value if it appears in this ranking (although Polygon PoS itself is a sidechain and excluded from growthepie's L2 universe — see [/answers/l2-vs-sidechain](/answers/l2-vs-sidechain)). The pattern is the same: when a token predates its associated L2 or serves a broader protocol, its market cap is not a clean measure of "the L2's economic value". Read the ranking with that asymmetry in mind. Tokens that are unambiguously L2-native (ARB → Arbitrum, OP → Optimism, STRK → Starknet, MNT → Mantle) don't carry this caveat. ### What's the difference between market cap and FDV? **Market cap** = circulating supply × current price. It only counts tokens already in circulation. **Fully-diluted valuation (FDV)** = total max supply × current price. FDV counts every token that will ever exist, including locked / vested / team / DAO-treasury tokens that haven't hit the market yet. **The MC / FDV ratio tells you what fraction of the supply is already in circulation** — a number close to 100% means most tokens are already trading; a low number (often 5–30% for newer L2 tokens) means a large portion of supply is still vested and could eventually enter the market. ### Which L2s have no native token? On 2026-05-27 UTC, the L2s in growthepie's tracked universe with **no native token** are: Arbitrum Nova, Base Chain, Ink, Soneium. The most prominent is **Base** — Coinbase has publicly stated they do not plan to issue a native token. Some others are early-stage chains that may launch tokens later. These chains are excluded from the market-cap ranking by definition, not because data is missing. ### What are the top 10 Ethereum L2 tokens by market cap? On 2026-05-27 UTC, the top 10 Ethereum L2 native tokens by USD market cap are: 1. Mantle (MNT) $2.16B (FDV $4.06B); 2. Unichain (UNI) $2.11B (FDV $3.32B); 3. World Chain (WLD) $1.13B (FDV $3.31B); 4. Arbitrum One (ARB) $676.81M (FDV $1.08B); 5. OP Mainnet (OP) $278.38M (FDV $556.33M); 6. Starknet (STRK) $257.88M (FDV $414.83M); 7. ZKsync Era (ZK) $143.83M (FDV $311.27M); 8. Linea $95.05M (FDV $226.69M); 9. MegaETH (MEGA) $83.64M (FDV $741.51M); 10. Plume Network (PLUME) $82.36M (FDV $143.15M). The list updates daily from growthepie's per-chain `/v1/metrics/chains/{chain}/market_cap.json` endpoints. ### What is the combined market cap of all Ethereum L2 tokens? Combined market cap across every tracked L2 token (data 2026-05-27 UTC): **$7.33B**. Combined FDV: **$14.75B**. The ratio gives the ecosystem-wide circulating supply share — typically lower than for older crypto markets because most L2 tokens are still vesting. ### Why does ARB lead the L2 token rankings? Arbitrum has had a mainnet token since the March 2023 airdrop, while many newer L2s launched tokens later or not at all. ARB benefits from being **the largest L2 by all-time TVL and one of the earliest production-grade rollups**, which translates into deeper liquidity, broader exchange listings, and a longer track record. The token serves DAO governance and ongoing emissions, with sequencer revenue and incentive programs flowing through Arbitrum's governance process. ### Why does Base not have a token if it's one of the biggest L2s? **Coinbase has publicly committed not to issue a Base token.** This is an intentional strategy choice — Coinbase, as a public company, derives revenue from Base's sequencer fees directly and prefers that over the complexity of running a token (which would invite securities-law scrutiny in the US and could fragment Coinbase's own revenue). For users, this means Base has no governance token to hold or stake; for developers it means no native-token incentive programs the way other L2s offer. Base's onchain activity numbers continue to lead despite the no-token strategy — see [/answers/base-vs-arbitrum](/answers/base-vs-arbitrum). ### How is the market cap calculated? **Market cap = circulating supply × current USD price.** growthepie pulls per-token market-cap and FDV time series from the per-chain endpoints (`/v1/metrics/chains/{chain}/market_cap.json` and `/v1/metrics/chains/{chain}/fdv.json`). The latest available daily row's USD value is the headline figure shown here. The endpoint sources its underlying price and supply data from standard market-data providers — so this page agrees directionally with CoinGecko / CoinMarketCap for the same tokens, though small differences in snapshot timing and circulating-supply definitions can produce minor variances. ### Is the ranking restricted to "true" L2s only? Yes — only chains in growthepie's curated Ethereum L2 universe are ranked here. That excludes sidechains (e.g. Polygon PoS — its token POL/MATIC has a large market cap but Polygon PoS is not an Ethereum L2 by the standard definition; see [/answers/l2-vs-sidechain](/answers/l2-vs-sidechain)) and Ethereum L1 itself. **25** L2s are in the universe on 2026-05-27 UTC; **21** of them have a native token tracked here; the remaining **4** are listed under "no native token" above. ### How does this compare to CoinGecko / CoinMarketCap? CoinGecko and CoinMarketCap rank tokens globally without distinguishing whether the token is for an Ethereum L2, an L1, a DeFi protocol, or something else. This page restricts the ranking to **Ethereum L2 native tokens only**, using growthepie's own L2 classification (the same one used across all other answer pages on this site). So this page answers "which L2 token has the highest market cap" specifically, not "which crypto token in general". Absolute MC values should agree closely with CoinGecko / CoinMarketCap for any given token. ### Why does this change rapidly over time? Two reasons: **(1) Price.** Token market cap is dominated by short-term price swings — a 20% price move shifts MC by 20% with no fundamental change. **(2) Supply unlocks.** L2 tokens with low circulating-share figures have large vesting / treasury supplies that unlock over months and years. When an unlock hits, circulating supply rises, increasing the MC even if price is flat. Treat any single snapshot as a daily picture, not a long-term valuation. ### Where can I see this updating live? growthepie's [economics dashboard](https://www.growthepie.com/economics) tracks per-chain market cap, FDV, fees, profit, and other token-relevant metrics live. Per-chain pages (e.g. [/chains/arbitrum](https://www.growthepie.com/chains/arbitrum), [/chains/optimism](https://www.growthepie.com/chains/optimism)) show each token in context with the chain's usage metrics — useful for "is this token expensive vs the activity that backs it" comparisons. ### Does a higher token market cap mean the chain is more important? No. Token market cap measures the speculative + governance value of one specific asset, not the strategic or technical importance of the chain. **Base has zero token market cap but leads in daily transactions, daily active addresses, and aggregate user activity.** Conversely, a chain with a high token MC but low onchain activity may have a token-trading market that's much larger than its actual economic activity. Use both lenses; see [/answers/most-used-ethereum-l2](/answers/most-used-ethereum-l2) for activity rankings and [/answers/most-value-secured-ethereum-l2](/answers/most-value-secured-ethereum-l2) for value-secured rankings. --- ## Are Ethereum L2 fees getting cheaper over time? URL: https://www.growthepie.com/answers/are-ethereum-l2-fees-getting-cheaper ### Direct answer Yes — Ethereum L2 user fees have gotten dramatically cheaper over time. Data 2026-05-27 UTC: the median L2 user fee in 2024-02 (pre-Dencun) was 20¢; in 2026-04 (latest available month) it is 0.06¢ — 307× cheaper. The single largest step-down was the Dencun upgrade in March 2024 (EIP-4844 blobs); Pectra (May 2025) and Fusaka (December 2025) compounded the effect. See the page for the per-upgrade anchor table. Live tracker: growthepie.com/fees. ### Full answer Short answer (data 2026-05-27 UTC): As of 2026-05-27 UTC, the median Ethereum L2 user fee is 0.06¢ per transaction (2026-04). Pre-Dencun (2024-02) the same median was 20¢ — fees are now 307× cheaper than before the March 2024 Dencun upgrade. Year-over-year (2025-04 → 2026-04): 2.03× cheaper (-50.7%). Updated daily — the trend on this page recomputes from growthepie's per-L2 monthly txcost endpoints. The in-progress current month is excluded so the latest figure shown is the most recent fully-completed month. The headline Cross-L2 median user fee, anchor months: - Pre-Dencun (2024-02) — 20¢ per L2 transaction. - Post-Dencun (2024-04) — 1.9¢. Change: 10.7× cheaper. - Pre-Pectra (2025-04) — 0.13¢. - Post-Pectra (2025-06) — 0.11¢. Change: 1.25× cheaper. - Pre-Fusaka (2025-11) — 0.10¢. - Post-Fusaka (2026-01) — 0.09¢. Change: 1.06× cheaper. - Latest available month (2026-04) — 0.06¢. Long-run change (2024-02 → 2026-04): 307× cheaper (-99.7%). Year-over-year change (2025-04 → 2026-04): 2.03× cheaper (-50.7%). What changed at each upgrade Dencun — March 13, 2024 (EIP-4844 blobs). Before Dencun, L2s posted their transaction data to Ethereum as calldata — expensive, because calldata wasn't purpose-built for L2 settlement. Dencun introduced blobs: a separate transaction type with its own fee market, designed specifically to carry L2 data. Per-byte L2 settlement cost dropped 10×. L2s passed the savings through to users; the cross-L2 median fee step-down between 2024-02 and 2024-04 was 10.7× cheaper. Pectra — May 7, 2025. Doubled the blob target from 3 to 6 per block. More headroom for L2 data without escalating the blob fee market. The 2025-04 → 2025-06 step was smaller than Dencun's because Pectra was a capacity expansion rather than a foundational architecture change — but it kept the curve moving down. See /answers/what-pectra-upgrade-changed. Fusaka — December 4, 2025. Added PeerDAS (EIP-7594) which lets nodes verify blob availability via sampling, enabling further blob-target increases. BPO1 + BPO2 raised the target to 14 blobs per block by January 2026. Also added EIP-7918, a floor on blob fees that prevented the "zero-fee blob" period (where blobs could trade for almost nothing, undermining L1 economics). The 2025-11 → 2026-01 step was 1.06× cheaper. See /answers/what-fusaka-upgrade-changed. Why fees fell so much Two compounding forces: 1. Protocol upgrades. Each upgrade above made the actual cost of L2 settlement cheaper — Dencun by introducing blobs, Pectra by raising capacity, Fusaka by raising capacity further and stabilising the blob fee market. The underlying input cost to running an L2 dropped 10× and then continued to drift down. 2. Competition. As more L2s launched, sequencers competed on user-facing fees to win activity. Chains that didn't pass through the DA savings lost users to chains that did. The market completed the work the protocol upgrades started. What this number does NOT measure A few honest framings to keep the trend in context: - Not the cheapest L2. The cross-L2 median is a typical-chain figure; some L2s are much cheaper, some more expensive. For the per-chain ranking see /answers/lowest-fee-ethereum-l2. - Not a specific transaction type. The median txcost averages across all transaction types on each chain — transfers, swaps, contract deployments, etc. Swaps cost more than transfers; the median falls in between. For swap-specific numbers see fees.growthepie.com. - Not Ethereum L1. This page is about L2 fees. Ethereum mainnet fees are a separate question — and L1 fees rose slightly after Dencun (Dencun increased blob revenue and reduced calldata revenue, the net effect on L1 depends on traffic mix). See /answers/ethereum-mainnet-revenue-from-l2s. - Not a per-transaction guarantee. Fees fluctuate with demand. The numbers here are monthly aggregates — a single transaction during a demand spike can cost much more than the headline median. Will L2 fees keep falling? Direction: probably down on average, but with bumps. Three things to track: - Future blob-target increases (BPO3 and beyond are on the Ethereum roadmap). More capacity → lower per-blob cost when supply exceeds demand. - L2 demand growth. If L2 activity grows faster than blob capacity, blob fees can rise and so can L2 settlement cost. - EIP-7918 blob fee floor (Fusaka). Prevented the "zero-fee blob" failure mode. Means fees won't go to zero even in low-demand periods — there's a structural minimum below which fees won't fall. Realistic outcome: the curve continues to drift down with cyclical bumps for capacity-vs-demand swings, not monotonic decline. Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue and filter to the curated L2 universe (bucket !== "Layer 1", deployment === "PROD", excludes sidechains and aggregate keys). 2. For each L2, pull and read (the period-native monthly USD value). 3. For each calendar month with data, compute the cross-L2 median across every L2 that has a value for that month — so each chain contributes one number per month regardless of activity volume. 4. Drop the trailing month if it's still in progress. 5. Look up anchor months by string: pre/post-Dencun (2024-02 / 2024-04), pre/post-Pectra (2025-04 / 2025-06), pre/post-Fusaka (2025-11 / 2026-01). 6. Compute per-upgrade and long-run multiples and percent changes. All values shown were generated on 2026-05-27 UTC. Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. The trend on this page is computed mechanically from public API data — chains and upgrades do not influence the median, and supporters do not receive any adjustments. Full list of supporters: growthepie.com/donate. Cross-check this answer. Compare the direction and the per-upgrade step-down magnitudes against L2BEAT, DefiLlama's fees view, and Etherscan's gas tracker for the L1 comparison. Absolute numbers will differ across providers (median vs average, all-tx vs swap-only, USD definitions), but the direction agrees. Related answers - /answers/lowest-fee-ethereum-l2 — per-chain ranking by current fee (the leaderboard view of the same data). - /answers/what-pectra-upgrade-changed and /answers/what-fusaka-upgrade-changed — detailed upgrade explainers. - /answers/ethereum-mainnet-revenue-from-l2s — the L1-revenue side of the same change. - /answers/what-is-data-availability — explains blobs and PeerDAS, the technical primitives that made this trend possible. ### FAQ ### Are Ethereum L2 fees getting cheaper over time? **Yes.** As of 2026-05-27 UTC, the median Ethereum L2 user fee is **0.06¢** per transaction (2026-04). Pre-Dencun (2024-02) the same median was 20¢ — fees are now **307× cheaper** than before the March 2024 Dencun upgrade. Year-over-year (2025-04 → 2026-04): 2.03× cheaper (-50.7%). The single biggest step-down was the **Dencun upgrade in March 2024** (EIP-4844 introduced blobs), which cut L2 settlement cost ~10× overnight. **Pectra (May 2025)** and **Fusaka (December 2025)** compounded the effect by raising blob capacity. Live tracker: [growthepie.com/fees](https://www.growthepie.com/fees). ### What was the L2 fee like before Dencun? In **2024-02** (the last full month before Dencun activated on 2024-03-13), the cross-L2 median user fee was **20¢** per transaction. That figure is the median across every tracked Ethereum L2's median monthly txcost — so it reflects what a typical L2 user typically paid, not a per-chain extreme. For context, transactions on Ethereum mainnet itself routinely cost dollars during the same period. ### How much did Dencun reduce L2 fees? Dencun activated on **2024-03-13** and introduced **blobs** (EIP-4844) — a new transaction type that gives L2s cheap data-availability. The effect on L2 fees was immediate and dramatic. **Pre-Dencun (2024-02): 20¢** median L2 fee per transaction. **Post-Dencun (2024-04): 1.9¢**. That's **10.7× cheaper** in the first full month after activation. Per-L2 settlement cost dropped roughly an order of magnitude, and L2s passed the savings through to users. See [/answers/what-pectra-upgrade-changed](/answers/what-pectra-upgrade-changed) for the upgrade context and [/answers/ethereum-mainnet-revenue-from-l2s](/answers/ethereum-mainnet-revenue-from-l2s) for the L1-revenue side of the same change. ### How much did Pectra (May 2025) change L2 fees? Pectra activated on **2025-05-07** and doubled the blob target from 3 to 6 per block — more headroom for L2 data without driving the blob fee market into escalation. The median L2 fee moved from **0.13¢** in 2025-04 to **0.11¢** in 2025-06 (1.25× cheaper). The Pectra step-down was smaller than Dencun's because Dencun was the foundational change — Pectra was a capacity expansion on top of an architecture that already worked. See [/answers/what-pectra-upgrade-changed](/answers/what-pectra-upgrade-changed). ### How much did Fusaka (December 2025) change L2 fees? Fusaka activated on **2025-12-04** with **PeerDAS** (EIP-7594) and a path to further blob target increases via BPO1 / BPO2 (target 6 → 14 per block by January 2026). The median L2 fee moved from **0.10¢** in 2025-11 to **0.09¢** in 2026-01 (1.06× cheaper). Fusaka also introduced **EIP-7918**, a floor on blob fees that prevented a "zero-fee blob" period. See [/answers/what-fusaka-upgrade-changed](/answers/what-fusaka-upgrade-changed). ### How much cheaper are L2 fees today than before Dencun? The cross-L2 median fee in **2024-02** (pre-Dencun) was **20¢**. In **2026-04** (latest available month) it is **0.06¢** — **307× cheaper** (-99.7%). The compound effect of Dencun + Pectra + Fusaka has cut a typical L2 transaction's user-paid cost by more than an order of magnitude in just over two years. ### How much have L2 fees changed in the past 12 months? Year-over-year: median L2 fee in **2025-04** was **0.13¢**, vs **0.06¢** in **2026-04** — fees are now cheaper by 2.03× cheaper (-50.7%). The 12-month window captures most of the Pectra + Fusaka step-downs. ### Why have L2 fees fallen so much? Two compounding reasons: **1. Cheaper data availability.** Pre-Dencun, L2s posted their transaction data as expensive Ethereum calldata. Dencun introduced **blobs** — a cheaper, purpose-built data type. Per-byte L2 settlement cost dropped ~10×, and Pectra + Fusaka raised blob capacity further so L2s have room to grow without per-blob fees escalating. **2. Competition.** As more L2s launched and bridge fragmentation became a real cost for users, chains competed on fees to win activity. Sequencers that don't pass through DA savings lose users to ones that do. The market does the rest of the work the protocol upgrades started. ### Will L2 fees keep falling? In the medium term, probably — additional blob target increases are planned (BPO2 → BPO3 schedule public on the [Ethereum roadmap](https://ethereum.org/en/roadmap/)), each of which raises L2 capacity without raising the per-blob price. **But the long-run direction depends on demand growth too.** If L2 demand grows faster than blob capacity, blob fees can rise and L2 settlement cost with them. **EIP-7918** (Fusaka) put a floor on blob fees to prevent the "zero-fee blob" failure mode — so fees won't go to zero even in low-demand periods. Realistic outcome: the curve continues to drift down with bumps for capacity-vs-demand swings, not monotonic decline. ### Are all L2 fees this low or are some chains more expensive? There's a wide spread across L2s. The median shown here is the cross-chain median of each L2's typical user fee — half the L2s are cheaper, half are more expensive. The cheapest L2s today are routinely sub-cent for simple transfers; some L2s with low blob-posting batching efficiency or higher overhead can be 5–10× the median. See [/answers/lowest-fee-ethereum-l2](/answers/lowest-fee-ethereum-l2) for the per-chain ranking and [growthepie.com/fees](https://www.growthepie.com/fees) for the live leaderboard. ### Do these numbers include swap fees or just transfers? These figures use growthepie's **median txcost** metric, which is the median USD fee across all transactions on a chain — not a single transaction type. Swap transactions are typically more expensive than transfers because they touch more contract logic, so the all-transactions median falls between a pure transfer cost and a pure swap cost. For swap-specific numbers see the [live fees view](https://fees.growthepie.com), which separates median, transfer, swap, and average. ### How is the cross-L2 median computed? For each completed month: pull the **monthly USD txcost** value from `/v1/metrics/chains/{chain}/txcosts.json` for every L2 in the curated universe. The cross-L2 median for that month is the median across those per-L2 values — so each chain contributes one number per month, no chain is over-weighted, and the median is robust to outliers. Monthly granularity (rather than daily) smooths out day-to-day volatility while still capturing the upgrade step-downs. The in-progress current month is excluded if it's not yet complete. ### Why use cross-chain median and not a weighted average? A volume-weighted average would tell you "the typical transaction's fee" if the goal is to estimate what a randomly-sampled L2 transaction cost. The cross-chain median tells you "what the typical L2 charges". The two answer different questions; the median is fairer for the "are L2s collectively getting cheaper" framing because it weights every L2 equally regardless of activity volume. (A high-volume cheap chain shouldn't make the headline number look more expensive than the typical chain actually was.) ### How many L2s contribute to the median each month? It grows over time. Early-2024 months have fewer chains (some L2s hadn't launched yet). Today, most L2s in growthepie's 25-chain curated universe contribute. The "Contributing chains" column in the monthly table on this page shows the per-month count so you can spot months with low-sample-size noise. ### Where can I see this data live? growthepie's [fees dashboard](https://www.growthepie.com/fees) is the live tracker — per-chain median, transfer, swap, and average user fees with hourly granularity. [fees.growthepie.com](https://fees.growthepie.com) is the simpler at-a-glance view. The historical per-chain series used by this page is at `/v1/metrics/chains/{chain}/txcosts.json`. ### How does this compare to other fee trackers? Cross-check the direction (and the step-down magnitudes around Dencun / Pectra / Fusaka) against [L2BEAT](https://l2beat.com) (per-chain fee tracking with its own methodology), [DefiLlama's fees view](https://defillama.com/fees), or Etherscan's gas tracker for L1 comparison. Absolute numbers will differ across providers because of definition choices (median vs average, all-tx vs swap-only, USD-at-execution vs USD-at-end-of-day), but the trend direction agrees. --- ## Which Ethereum L2 is growing the fastest? URL: https://www.growthepie.com/answers/fastest-growing-ethereum-l2 ### Direct answer By month-over-month and quarter-over-quarter growth across throughput, transactions, active addresses, and total value secured: month-over-month: Taiko Alethia leads throughput growth (+139%), Taiko Alethia leads transactions (+161%), Manta Pacific leads active addresses (+143%), Ronin leads TVS (+5911%). quarter-over-quarter: MegaETH leads throughput growth (+4667%), MegaETH leads transactions (+72441%), MegaETH leads active addresses (+38205%), Ronin leads TVS (+7420%). Data: 2026-05-27 UTC. Live leaderboards: growthepie.com/fundamentals/throughput. ### Full answer Short answer (data 2026-05-27 UTC): Month-over-month, the fastest-growing Ethereum L2s are Taiko Alethia (+139%; 0.07 Mgas/s now vs 0.03 Mgas/s prior) by throughput, Taiko Alethia (+161%; 1.68M now vs 642.6k prior) by transactions, Manta Pacific (+143%; 21.0k now vs 8.6k prior) by active addresses, and Ronin (+5911%; $265.71M now vs $4.42M prior) by total value secured (TVS). Quarter-over-quarter leaders may differ — see the per-metric tables below. Top 3 by metric and timeframe Throughput growth (Mgas/s) - Month-over-month: 1. Taiko Alethia (+139%; 0.07 Mgas/s now vs 0.03 Mgas/s prior); 2. Lisk (+8.7%; 0.09 Mgas/s now vs 0.08 Mgas/s prior); 3. MegaETH (+6.9%; 6.96 Mgas/s now vs 6.51 Mgas/s prior). - Quarter-over-quarter: 1. MegaETH (+4667%; 100.5 Mgas/s now vs 2.11 Mgas/s prior); 2. OP Mainnet (+20.8%; 9.25 Mgas/s now vs 7.66 Mgas/s prior); 3. Base Chain (+6.2%; 29.5 Mgas/s now vs 27.8 Mgas/s prior). Transaction count growth - Month-over-month: 1. Taiko Alethia (+161%; 1.68M now vs 642.6k prior); 2. Lisk (+14.2%; 279.4k now vs 244.5k prior); 3. Mantle (+12.2%; 1.23M now vs 1.10M prior). - Quarter-over-quarter: 1. MegaETH (+72441%; 11.41B now vs 15.73M prior); 2. OP Mainnet (+18.6%; 193.01M now vs 162.79M prior); 3. Scroll (+7.5%; 7.52M now vs 6.99M prior). Active addresses growth - Month-over-month: 1. Manta Pacific (+143%; 21.0k now vs 8.6k prior); 2. Soneium (+108%; 196.4k now vs 94.4k prior); 3. Arbitrum One (+31.5%; 3.30M now vs 2.51M prior). - Quarter-over-quarter: 1. MegaETH (+38205%; 488.8k now vs 1.3k prior); 2. Ink (+297%; 629.7k now vs 158.7k prior); 3. Celo (+24.4%; 3.32M now vs 2.67M prior). Total value secured (TVS) growth - Month-over-month: 1. Ronin (+5911%; $265.71M now vs $4.42M prior); 2. MegaETH (+139%; $261.21M now vs $109.13M prior); 3. World Chain (+19.5%; $330.09M now vs $276.32M prior). - Quarter-over-quarter: 1. Ronin (+7420%; $265.71M now vs $3.53M prior); 2. MegaETH (+109%; $261.21M now vs $125.28M prior); 3. Plume Network (+36.4%; $88.68M now vs $65.02M prior). Updated daily — every growth ranking on this page is recomputed from growthepie's public per-chain timeseries. Current = most recent completed period; we deliberately skip the in-progress period to keep the rankings stable day-to-day. How we measure growth We compare each L2's last completed period to the period immediately before it, for four metrics: - Throughput (Mgas/s) — hardest to game. A chain can't inflate gas-per-second with cheap spam because every operation costs gas proportional to its complexity. - Transaction count — most intuitive. Sensitive to fee reductions and hot apps launching. - Active addresses — closest proxy for "real users". Period-native unique counts mean a multi-day user is counted once per window. - Total value secured (TVS) — the dollar value of assets bridged to or natively held on the chain. Capital-weight signal; growth here reflects new deposits net of withdrawals. Two windows: - Month-over-month — last completed month vs the month before. Responsive. - Quarter-over-quarter — last completed quarter vs the quarter before. Smooths out monthly volatility. Minimum-activity filter. Tiny chains with small baselines can post implausible growth percentages (100 → 1000 addresses reads as +900%). To keep the rankings meaningful, we require each chain to clear a minimum CURRENT-period value before being eligible: ≥0.05 Mgas/s throughput, ≥50k transactions per month, ≥1,000 unique addresses per month, ≥$10M TVS. Adjust the thresholds and you'd get a different list — what we publish is calibrated to surface chains people recognise. Fastest-growing L2s — month-over-month As of 2026-05-27 UTC (month-over-month), the fastest-growing Ethereum L2s are Taiko Alethia by throughput (+139%), Taiko Alethia by transactions (+161%), Manta Pacific by active addresses (+143%), Ronin by TVS (+5911%). Fastest-growing L2s — quarter-over-quarter As of 2026-05-27 UTC (quarter-over-quarter), the fastest-growing Ethereum L2s are MegaETH by throughput (+4667%), MegaETH by transactions (+72441%), MegaETH by active addresses (+38205%), Ronin by TVS (+7420%). Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue and filter to chains where , , and the chain key is not on the explicit non-L2 list below. 2. For each L2 and each metric, pull the per-chain timeseries endpoint ( ). 3. For flow metrics (throughput, txcount, daa): read and . Take the second-to-last entry as the current value (most recent completed period) and the third-to-last as the prior value. 4. For TVS (a stock metric): the per-chain TVL endpoint only exposes a daily series, so read , resolve the USD column by name, and sample the latest day plus offsets at −30 days (monthly prior) and −90 days (quarterly prior). 5. Compute growth as . Skip chains whose current value is below the minimum-activity threshold for that metric. 6. Sort chains by growth %, descending. Take the top 10 per (metric, window) pair. All values shown on this page were generated on 2026-05-27 UTC from growthepie's public API: - Master chain list (with bucket / chaintype classification): - Per-chain throughput timeseries: - Per-chain transaction-count timeseries: - Per-chain active-address timeseries: - Per-chain TVL / TVS timeseries (daily, USD): Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Some supporters operate chains that appear in the rankings above. Rankings are computed mechanically from public API data — chains do not pay for inclusion or placement, and supporters do not receive ranking adjustments or preferential treatment. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. Independent L2 growth sources include L2BEAT's activity view (per-chain transaction counts over time), Dune Analytics (community-built growth dashboards), and the chains' own community reports. Definitions differ — L2BEAT and growthepie can disagree on which chains count as L2s, and growth windows differ between providers. When rankings disagree, comparing the underlying chain inclusion lists and window definitions is usually more informative than the ranks themselves. Which chains are included? The list of 25 chains is computed automatically from and refreshed when growthepie adds or removes coverage: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Ethereum mainnet — it is Layer 1, not Layer 2. - Polygon PoS — a sidechain with its own validator set, not a Layer 2. - Aggregate keys ( , ) — not individual chains. - Chains below the minimum-activity threshold — see methodology above. Excluding tiny chains from growth rankings is a deliberate choice, not an oversight. ### FAQ ### Which Ethereum L2 is growing the fastest? It depends on which metric and which window. As of 2026-05-27 UTC, **month-over-month** growth leaders: Taiko Alethia (+139%; 0.07 Mgas/s now vs 0.03 Mgas/s prior) by throughput; Taiko Alethia (+161%; 1.68M now vs 642.6k prior) by transactions; Manta Pacific (+143%; 21.0k now vs 8.6k prior) by active addresses; Ronin (+5911%; $265.71M now vs $4.42M prior) by total value secured (TVS). **Quarter-over-quarter** growth leaders may differ — see the per-metric tables below. Each ranking filters for minimum current-period activity so a brand-new chain with a tiny baseline doesn't claim implausible growth. Live leaderboards: [growthepie.com/fundamentals/throughput](https://www.growthepie.com/fundamentals/throughput). ### How is "growing fastest" measured? We compare each L2's **last completed period** to the **period immediately before it**, for each of four metrics. Monthly = last completed month vs the month before. Quarterly = last completed quarter vs the quarter before. Growth is `(current − prior) / prior`, expressed as a percentage. The three flow metrics (throughput, transactions, active addresses) use growthepie's per-chain period-native timeseries, so we don't fake a rolling window by summing dailies — which matters for active addresses, where summing dailies would double-count multi-day users. **TVS is a stock metric**, so it's sampled point-in-time: latest day vs 30 days ago (monthly) or 90 days ago (quarterly). ### Which Ethereum L2 has the fastest throughput growth month-over-month? On 2026-05-27 UTC, the top three Ethereum L2s by month-over-month throughput growth are 1. Taiko Alethia (+139%; 0.07 Mgas/s now vs 0.03 Mgas/s prior); 2. Lisk (+8.7%; 0.09 Mgas/s now vs 0.08 Mgas/s prior); 3. MegaETH (+6.9%; 6.96 Mgas/s now vs 6.51 Mgas/s prior). Throughput growth is the hardest of the three metrics to game — a chain can't inflate gas-per-second with cheap spam. ### Which Ethereum L2 has the fastest transaction growth month-over-month? On 2026-05-27 UTC, the top three Ethereum L2s by month-over-month transaction growth are 1. Taiko Alethia (+161%; 1.68M now vs 642.6k prior); 2. Lisk (+14.2%; 279.4k now vs 244.5k prior); 3. Mantle (+12.2%; 1.23M now vs 1.10M prior). Transaction growth often reflects fee reductions, hot apps launching, or airdrop farming — see the cross-check guidance below. ### Which Ethereum L2 has the fastest active-address growth month-over-month? On 2026-05-27 UTC, the top three Ethereum L2s by month-over-month active-address growth are 1. Manta Pacific (+143%; 21.0k now vs 8.6k prior); 2. Soneium (+108%; 196.4k now vs 94.4k prior); 3. Arbitrum One (+31.5%; 3.30M now vs 2.51M prior). Active addresses are unique per period — a wallet active on multiple days within the month is counted once, so this isn't inflated by per-user activity bursts the way summed-daily counts would be. ### Which Ethereum L2 has the fastest total-value-secured (TVS) growth month-over-month? On 2026-05-27 UTC, the top three Ethereum L2s by month-over-month TVS growth are 1. Ronin (+5911%; $265.71M now vs $4.42M prior); 2. MegaETH (+139%; $261.21M now vs $109.13M prior); 3. World Chain (+19.5%; $330.09M now vs $276.32M prior). TVS measures the dollar value of assets bridged to or natively held on the chain — growth here reflects net deposits over the window. Sampled point-in-time (today vs 30 days ago) since TVS is a stock, not a flow. ### Which Ethereum L2 has the fastest throughput growth quarter-over-quarter? On 2026-05-27 UTC, the top three Ethereum L2s by quarter-over-quarter throughput growth are 1. MegaETH (+4667%; 100.5 Mgas/s now vs 2.11 Mgas/s prior); 2. OP Mainnet (+20.8%; 9.25 Mgas/s now vs 7.66 Mgas/s prior); 3. Base Chain (+6.2%; 29.5 Mgas/s now vs 27.8 Mgas/s prior). Quarter-over-quarter smooths out monthly volatility — a chain that posts +200% one month and −80% the next won't lead this list, but a chain with steady multi-month growth will. ### Which Ethereum L2 has the fastest transaction growth quarter-over-quarter? On 2026-05-27 UTC, the top three Ethereum L2s by quarter-over-quarter transaction growth are 1. MegaETH (+72441%; 11.41B now vs 15.73M prior); 2. OP Mainnet (+18.6%; 193.01M now vs 162.79M prior); 3. Scroll (+7.5%; 7.52M now vs 6.99M prior). ### Which Ethereum L2 has the fastest active-address growth quarter-over-quarter? On 2026-05-27 UTC, the top three Ethereum L2s by quarter-over-quarter active-address growth are 1. MegaETH (+38205%; 488.8k now vs 1.3k prior); 2. Ink (+297%; 629.7k now vs 158.7k prior); 3. Celo (+24.4%; 3.32M now vs 2.67M prior). ### Which Ethereum L2 has the fastest TVS growth quarter-over-quarter? On 2026-05-27 UTC, the top three Ethereum L2s by quarter-over-quarter TVS growth are 1. Ronin (+7420%; $265.71M now vs $3.53M prior); 2. MegaETH (+109%; $261.21M now vs $125.28M prior); 3. Plume Network (+36.4%; $88.68M now vs $65.02M prior). Quarter-over-quarter TVS samples the same point-in-time on the latest day vs 90 days ago — smoother than monthly because short-term bridge flows wash out. ### Why exclude tiny chains from the growth ranking? Because tiny baselines produce silly numbers. A new L2 that goes from 100 daily transactions to 1,000 is growing 900%, but it's not meaningfully scaling Ethereum — and including it would push every chain that actually matters out of the top 10. We apply a minimum-activity threshold on the CURRENT period (e.g. ≥0.05 Mgas/s of throughput, ≥50k transactions per month, ≥1,000 unique addresses per month) before ranking by growth %. Adjust the threshold and you'd get a different ranking — what we publish is the threshold that empirically produces a list of chains people recognise as meaningful. ### What does "current period" mean exactly? The most recent **completed** period. Monthly = the last calendar month that has fully closed. Quarterly = the last quarter that has fully closed. We deliberately avoid the in-progress period — its values change every day and would make the rankings flicker. As a result the "current" period is usually 1–30 days old depending on when in the month the page is fetched. ### Is Polygon PoS counted as an L2 here? No. Polygon PoS is a sidechain with its own validator set and is excluded from these growth rankings, matching the rest of the answer pages on growthepie. Polygon zkEVM is a ZK rollup and is included. ### How many L2s are included? 25 chains. The full list (computed on 2026-05-27 UTC from growthepie's master chain catalogue) is: arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. A chain only appears in a given ranking if it has data for at least two completed periods AND its current-period value clears the minimum-activity threshold for that metric. ### Where does this answer come from? Per-chain values come from growthepie's per-chain timeseries endpoints (`/v1/metrics/chains/{chain}/{metric}.json`). For the three flow metrics (throughput, txcount, daa) we read `details.timeseries.monthly.data` and `.quarterly.data` and take the most-recent-completed and prior entries. For TVS we read `details.timeseries.daily.data` (the per-chain TVL endpoint only exposes a daily series) and sample the USD column at the latest day plus offsets of 30 days (monthly prior) and 90 days (quarterly prior) — point-in-time comparison appropriate for a stock metric. Growth is `(current − prior) / prior`. L2 membership comes from `master.json` (chains where `bucket !== "Layer 1"` and `chain_type` indicates an Ethereum rollup or validium). Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. No editorial overrides. ### Why does the fastest-growing chain not match the "most used" chain? Because they answer different questions. "Most used" measures the absolute level of activity right now — which chain processes the most transactions, has the most users, the most throughput. "Fastest-growing" measures the rate of change — which chain is increasing fastest from period to period. A mature chain like Base or Arbitrum tops most-used rankings but rarely tops growth rankings; the growth rankings are usually led by smaller chains that recently launched a hot app, reduced fees, or won a major migration. ### Can a chain be growing fast on one metric but flat on another? Absolutely, and the page intentionally exposes all three. A chain that adds a token-launch platform may post massive transaction growth without much throughput growth (the transactions are cheap mints). A chain that wins a single high-value app may grow throughput without growing addresses. A chain that runs an airdrop campaign may grow addresses without sustained transaction growth. We rank all three metrics independently so an honest reader can see where the growth is concentrated. ### Where can I see live L2 growth data? growthepie tracks every L2 with daily, weekly, monthly, and quarterly timeseries at [growthepie.com/fundamentals/throughput](https://www.growthepie.com/fundamentals/throughput), [/transaction-count](https://www.growthepie.com/fundamentals/transaction-count), and [/daily-active-addresses](https://www.growthepie.com/fundamentals/daily-active-addresses). Per-chain pages (e.g. [/chains/base](https://www.growthepie.com/chains/base)) show period-over-period change inline. ### How is "Ethereum L2" defined here? An Ethereum Layer 2 is a chain that derives security from Ethereum by posting transaction data and/or state to Ethereum mainnet. This includes optimistic rollups, ZK rollups, and Validiums. Sidechains (independent validator sets, like Polygon PoS) are excluded. --- ## What did Ethereum's "Fusaka" upgrade change? URL: https://www.growthepie.com/answers/what-fusaka-upgrade-changed ### Direct answer Fusaka activated on December 3rd, 2025 with 12 EIPs across three goals: scale blob capacity (PeerDAS + Blob-Parameter-Only upgrades tripled the blob target from 6 to 14 per block by January 2026), scale L1 gas (EIP-7935 raised the default block gas limit to 60M), and sustainable blob economics (EIP-7918 introduced a floor under blob fees, ending the "zero-fee" era). The full impact is tracked live at growthepie.com/quick-bites/fusaka. ### Full answer Short answer: Fusaka activated on December 3rd, 2025 and shipped 12 EIPs across three goals: (1) scale blob capacity through PeerDAS (EIP-7594) and the Blob-Parameter-Only framework (EIP-7892), which tripled the blob target from 6 → 14 per block via two follow-up BPO upgrades; (2) scale L1 gas (EIP-7935 raised the default block gas limit to 60M with a 30M target); and (3) sustainable blob economics (EIP-7918 puts a floor under blob fees so the 'zero-fee era' ends). Live impact so far — 19,441,656 blobs posted and Ξ4,733.33 in blob fees paid since Dencun. The full data tracker with live charts is at growthepie.com/quick-bites/fusaka. The summary below pulls the headline facts from the same source. Timeline - Dec 3, 2025 — Fusaka activates on Ethereum mainnet with 12 EIPs. - Dec 9, 2025 — BPO1: blob target raised from 6 → 10 per block (max 9 → 15). - Jan 7, 2026 — BPO2: blob target raised from 10 → 14 per block (max 15 → 21). BPOs (Blob-Parameter-Only upgrades) are lightweight protocol updates — they only change blob parameters, so they don't need a full hard fork each time. 1. Scale blob capacity Fusaka roughly tripled Ethereum's data availability ceiling. Two changes work together: - EIP-7594 (PeerDAS) — Peer Data Availability Sampling lets nodes verify blob availability without downloading every blob in full. Each node samples a small random fraction of each blob and gossips its samples; collectively the network confirms the blob is available even though no single node has all of it. - EIP-7892 (BPO framework) — Blob-Parameter-Only upgrades. A way to raise blob targets between major hard forks. BPO1 (Dec 9, 2025) took target from 6 → 10; BPO2 (Jan 7, 2026) took it from 10 → 14. These matter because every L2 rollup posts its transaction data to Ethereum as blobs. More blob capacity = either lower L2 fees, more L2 transactions, or both. 2. Scale L1 gas EIP-7935 set the default L1 block gas limit to 60M (with a 30M target). In practice the network reached this level shortly before Fusaka itself via validator signaling — Ethereum allows gradual gas-limit adjustments outside hard-fork cycles. The headroom matters most during peak demand: when L1 transaction volume runs into the old ceiling, the higher limit lets more transactions clear per block without driving fees up. 3. Sustainable blob economics (EIP-7918) EIP-7918 introduces a floor on blob gas price by bounding the blob base fee to execution gas prices. The exact rule: the cost of of blob gas can't fall below the equivalent execution cost ( ). When the floor binds, excess blob gas accumulates without subtracting the target — which tightens the blob fee market and anchors blob pricing to L1 execution economics. Why it matters: - Blob fees previously collapsed to effectively zero during periods of low rollup activity. That was great for L2 users but bad for Ethereum's long-term economic sustainability. - A permanent zero-fee floor also makes blob spam cheap — security implications, not just economic. - Anchoring blob pricing to L1 execution means blob fees now track the broader Ethereum fee market instead of collapsing in isolation. Live impact since Fusaka: Ξ16.51 in blob fees under EIP-7918, vs Ξ0.000005266 under the legacy path — a 3,134,728× multiplier. Verify on the Fusaka tracker. How does Fusaka compare to Dencun and Pectra? - Dencun (March 2024) — introduced blobs (EIP-4844 / proto-danksharding). Target 3 per block, max 6. - Pectra (May 2025) — doubled blob capacity. Target 6, max 9. - Fusaka (December 2025 + BPOs through January 2026) — roughly tripled blob capacity again to target 14, max 21. Added PeerDAS so further increases are safer. Added EIP-7918 so blob economics tighten. Short version: Dencun = blobs exist; Pectra = blobs scale 2×; Fusaka = blobs scale 5× from Dencun and the fee market matures. Methodology and data sources All facts on this page come from two sources: 1. growthepie's Fusaka data tracker — the live KPIs quoted in this page (total blobs, blob fees, EIP-7918 multipliers) are served by and and refresh daily from chain RPC data. 2. Ethereum's published EIP repository and upgrade announcement — for upgrade dates, EIP numbers, and parameter values (60M gas limit, 14 blob target, etc.). No editorial interpretation: every claim on this page is verifiable against the tracker or the EIP text. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Fusaka is an Ethereum mainnet upgrade — no L2 chain operator (whether or not a growthepie supporter) influences the upgrade or how it's described here. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. For the canonical Ethereum-side documentation see ethereum.org's Fusaka page, the EIP index for individual EIP text, and the ethereum/EIPs GitHub repo for the discussion threads behind each change. The growthepie Fusaka tracker linked above quantifies the upgrade's impact on the chain in the days and weeks after activation. ### FAQ ### What did Ethereum's "Fusaka" upgrade change? Fusaka activated on **December 3rd, 2025** and bundled 12 EIPs across three goals: **scale blob capacity** (EIP-7594 "PeerDAS" + EIP-7892 "Blob-Parameter-Only" upgrades), **scale L1 gas** (EIP-7935 raised the default block gas limit to 60M with a 30M target), and **sustainable blob economics** (EIP-7918 puts a floor under blob fees so the "zero-fee" blob era ends). Two follow-up BPO upgrades activated soon after — **BPO1** on December 9th, 2025 (target 6→10 blobs/block, max 9→15) and **BPO2** on January 7th, 2026 (target 10→14, max 15→21). Since Dencun, Ethereum has posted **** blobs paying **Ξ** in fees. ### When did Fusaka activate? Fusaka activated on **December 3rd, 2025**. Two follow-up Blob-Parameter-Only (BPO) upgrades followed: **BPO1 on December 9th, 2025** and **BPO2 on January 7th, 2026**. BPOs are lightweight protocol updates that only change blob-related parameters — they don't require a full hard fork. ### How many EIPs did Fusaka include? Twelve. The headline ones are EIP-7594 (PeerDAS), EIP-7892 (Blob-Parameter-Only upgrade framework), EIP-7935 (raised the L1 block gas limit to 60M default), and EIP-7918 (blob fee floor). See growthepie's [Fusaka tracker](/quick-bites/fusaka) for live data on what each one is doing in practice. ### What is PeerDAS (EIP-7594)? **Peer Data Availability Sampling** — the protocol change that lets Ethereum nodes verify blob availability *without downloading every blob in full*. Each node downloads only a small randomly-sampled fraction of each blob and gossips its samples to peers; collectively the network can confirm that the blob is fully available even though no single node has all of it. This is what makes raising the blob target safe — without PeerDAS, every full node would have to download every blob, and the data load would saturate consumer-grade nodes. ### What are Blob-Parameter-Only (BPO) upgrades? BPOs (EIP-7892) are lightweight protocol updates that adjust blob-related parameters without introducing broader execution or consensus changes. The point is to scale blob capacity incrementally without needing a full hard fork each time — Ethereum can now raise blob targets between major upgrades. **BPO1** activated December 9th, 2025 (target 6→10 blobs/block, max 9→15). **BPO2** activated January 7th, 2026 (target 10→14, max 15→21). ### Why does increasing blob capacity matter for L2s? Blobs are how Ethereum rollups post their transaction data to L1 for data availability. More blob capacity = more transaction data L2s can post per block = either lower fees for the same volume of L2 transactions, or the same fees with more L2 transactions, or some mix of both. Fusaka roughly tripled the blob target through Fusaka + BPO1 + BPO2 (from 6 → 14 per block), substantially expanding the headroom for L2 throughput. ### How much did Fusaka scale Ethereum L1 gas? EIP-7935 set the **default L1 block gas limit to 60M** (with a 30M target). In practice the network reached this level shortly before the hard fork via validator signaling — Ethereum allows gradual gas-limit adjustments independent of fork activation. The net effect is more L1 execution capacity, which lowers the marginal cost of L1 gas under load. ### Does Fusaka lower L1 gas fees? It removes a capacity ceiling, which lowers fees *when L1 demand exceeds the old ceiling*. When demand is below capacity the gas market is already in equilibrium — additional headroom doesn't reduce fees further. The Fusaka tracker has [live L1 gas charts](/quick-bites/fusaka) comparing average gas used vs the gas limit over time. ### What does EIP-7918 do? EIP-7918 introduces a **floor on blob gas price** by bounding the blob base fee to execution gas prices. Specifically, the cost of `GAS_PER_BLOB` of blob gas can't fall below the equivalent execution cost (`BLOB_BASE_COST × base_fee_per_gas`). When this floor binds, excess blob gas accumulates without subtracting the target, tightening the blob fee market and anchoring blob pricing to L1 execution economics. The practical effect: the "zero-fee blob era" — periods when blob fees collapsed to effectively nothing — is over. Since Fusaka, EIP-7918 has captured **Ξ** in blob fees vs **Ξ** under the old path — a **× multiplier**. ### Why does Ethereum want blob fees to have a floor? Two reasons. First, **economic sustainability** — blobs are part of Ethereum's value capture, and a market that frequently collapses to near-zero fees can't monetise data availability long-term. Second, **DoS protection** — a permanent near-zero blob fee makes it cheap to spam blob space, eroding the security model that L2s rely on. EIP-7918 ties blob pricing to L1 execution economics so blob fees track the broader Ethereum fee market instead of collapsing in isolation. ### What changed for end users? Indirectly, a lot. **L2 fees got cheaper** because L2s have more blob capacity to post into. **L1 transactions can run a bit more cheaply at peak times** because the L1 gas limit is higher. **Blob fees are slightly higher** during quiet periods because of EIP-7918's floor (though absolute values are still tiny). Users don't see Fusaka mechanics directly — they see the downstream effects on whichever L2 or app they use. ### Did Fusaka break anything for existing apps? No. Fusaka is a non-breaking upgrade — existing smart contracts, wallets, and L2 implementations continue to work without changes. The relevant changes are validator-side (PeerDAS sampling), protocol-side (blob parameter changes via BPOs), and economic (the EIP-7918 floor). Apps don't need to migrate. ### How does Fusaka compare to Dencun and Pectra? Dencun (March 2024) **introduced blobs** (EIP-4844 / proto-danksharding) with a target of 3 / max 6 per block — the foundation of modern L2 scaling. Pectra (May 2025) **doubled blob capacity** to a target of 6 / max 9 per block. Fusaka (December 2025) **tripled it again** to a target of 14 / max 21 by the time BPO2 landed in January 2026, and added PeerDAS so further increases are safer. So roughly: Dencun = blobs exist; Pectra = blobs scale 2×; Fusaka = blobs scale ~5× and the economics tighten. ### How is Fusaka tracked on growthepie? See the [Fusaka tracker](/quick-bites/fusaka) for live charts: blob count vs target since each upgrade, L1 gas used vs limit, blob fees vs L1 gas fees as a share of total ETH burn, and a comparison of blob fees under EIP-7918 vs the legacy path. The tracker is what backs the live numbers quoted on this answer page. ### Where do these numbers come from? Live KPIs (, , EIP-7918 multipliers) come from growthepie's Fusaka data endpoints (`/v1/quick-bites/fusaka/totals.json` and `/v1/quick-bites/fusaka/eip7918_kpis.json`), which are refreshed daily from chain RPC data. Upgrade dates, EIP numbers, and parameter values come from the published Ethereum upgrade announcement and EIP repository. No editorial interpretation — every claim on this page is verifiable against the linked tracker or the EIP text. ### What's next after Fusaka? Two threads. Further **BPO upgrades** can raise blob targets again without a full hard fork — the framework introduced in Fusaka makes this routine. And **Glamsterdam**, Ethereum's next major scheduled upgrade, will introduce additional improvements to data availability and execution. Check [the Ethereum roadmap](https://ethereum.org/en/roadmap/) for the latest scheduled changes. --- ## What did Ethereum's "Pectra" upgrade change? URL: https://www.growthepie.com/answers/what-pectra-upgrade-changed ### Direct answer Pectra activated on May 7th, 2025 with 11 EIPs across three goals: doubling blob capacity (EIP-7691 raised the blob target from 3 to 6 per block, halving L2 fees), introducing smart-account features for ordinary wallets (EIP-7702 added "Set Code" / Type 4 transactions enabling gas sponsorship, non-ETH gas payment, and transaction batching), and improving staking (EIP-7251 raised the validator effective stake cap from 32 ETH to 2,048 ETH so rewards can compound; EIP-7002 and EIP-6110 simplified the deposit and withdrawal pipeline). Pectra is widely called the "user experience" upgrade. Live adoption: growthepie.com/quick-bites/pectra-upgrade. ### Full answer Short answer: Pectra activated on May 7th, 2025 and shipped 11 EIPs across three user-facing goals: (1) more blobs (EIP-7691 doubled the blob target from 3 to 6 per block, max 6→9 — cheaper L2 transactions), (2) smarter wallets (EIP-7702 lets ordinary wallets temporarily act as smart contract accounts via a new 'Set Code' / Type 4 transaction type — gas sponsorship, non-ETH gas, transaction batching), and (3) staking upgrades (EIP-7251 raised the validator effective stake cap from 32 ETH to 2,048 ETH so rewards compound; EIP-7002 and EIP-6110 simplified the deposit / withdrawal pipeline). Live adoption data: growthepie.com/quick-bites/pectra-upgrade. Less clicks, less signatures, more blobs. Past Ethereum upgrades focused on technical improvements; Pectra was the first one where user-experience features dominated the announcement. Timeline - May 7, 2025 — Pectra activates on Ethereum mainnet. The fork is the combined "Prague" execution-layer + "Electra" consensus-layer hard fork. - December 3, 2025 — Fusaka follows, tripling blob capacity again (to a target of 14 by BPO2 in January 2026). See /answers/what-fusaka-upgrade-changed. 1. More blobs (EIP-7691) Rollups had been operating at full blob capacity for months before Pectra — the blob market was perpetually saturated, driving up L2 fees. EIP-7691 fixed that immediately: - Blob target per block: doubled from 3 → 6. - Blob maximum per block: raised from 6 → 9. - Net effect: roughly halved blob fees the moment Pectra activated, with the savings flowing through to L2 users in the following weeks. The blob-fee market only escalates via EIP-1559 once usage exceeds the target, so the extra headroom also kept fees lower during demand spikes. See the live "blobs per block vs target" chart on the Pectra tracker. 2. Smarter wallets (EIP-7702) EIP-7702 introduces a new transaction format — "Set Code" or Type 4 — that lets a regular wallet (EOA — externally-owned account) temporarily act as a smart contract account for one transaction. This unlocks features that previously required dedicated smart-contract wallets: - Gas sponsorship. Someone else (a dApp, a custodian) pays the gas fee. - Pay gas in non-ETH tokens. Pay fees in USDC, a project's native token, or anything else the wallet supports. - Transaction batching. Multiple actions in one user-signed transaction — no more "approve, then swap" two-step. - Session keys / delegation. A dApp can sign on the user's behalf within preset limits. Critically, users don't have to switch wallets to use EIP-7702 features — their existing EOA can opt in by signing a Set Code authorization. Adoption is gated by wallet support: a wallet has to expose the Type 4 transaction format to its users for any of this to surface. See the live "Set Code transactions per chain" chart on the Pectra tracker for adoption across Ethereum L1, Base, OP Mainnet, Unichain, and Arbitrum. 3. Staking upgrades (EIP-7251, 7002, 6110) Three EIPs together simplified the staking lifecycle: - EIP-7251 — MaxEB. Raised the maximum effective stake per validator from 32 ETH to 2,048 ETH. Large stakers (custodians, exchanges, big solo stakers) can now consolidate many validators into one and let rewards compound rather than spinning up new validators every time the balance accrues above 32 ETH. Small solo stakers can still run 32 ETH validators — the change is opt-in. - EIP-7002 — Execution-layer-triggered exits and withdrawals. Validators can now trigger their own exits and partial withdrawals from the execution layer instead of requiring a separate withdrawal credential setup. - EIP-6110 — On-chain deposits. Moved the deposit pipeline onto the execution layer, removing the multi-hour deposit delay that previously existed. Side benefit: fewer validators on the network (because large stakers can now use one validator instead of N) reduces consensus-layer load and makes future consensus upgrades easier. How does Pectra compare to Dencun and Fusaka? - Dencun (March 2024) — introduced blobs (EIP-4844 / proto-danksharding). Target 3 per block, max 6. - Pectra (May 2025) — doubled blob capacity (target 6, max 9) and introduced smart-account features via EIP-7702. The "user experience" upgrade. - Fusaka (December 2025 + BPOs through January 2026) — tripled blob capacity again (target 14, max 21). Added PeerDAS for safer blob scaling. Added EIP-7918 blob fee floor. Short version: Dencun = blobs exist; Pectra = blobs scale 2× and wallets get smarter; Fusaka = blobs scale 5× from Dencun and fee economics tighten. Methodology and data sources All facts on this page come from two sources: 1. growthepie's Pectra tracker — live adoption charts (blob count vs target on Ethereum L1, Type 4 transactions per chain) served by and refreshed daily. 2. Ethereum's published EIP repository and Pectra announcement — for upgrade dates, EIP numbers, parameter values (blob target = 6, validator cap = 2,048 ETH). No editorial interpretation: every claim on this page is verifiable against the tracker or the EIP text. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Pectra is an Ethereum mainnet upgrade — no L2 chain operator (whether or not a growthepie supporter) influences the upgrade or how it's described here. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. For canonical documentation see ethereum.org's Pectra page, the EIP index for individual EIP text, and the ethereum/EIPs GitHub repo for the discussion threads behind each change. The growthepie Pectra tracker linked above quantifies the upgrade's adoption in the weeks and months after activation. ### FAQ ### What did Ethereum's "Pectra" upgrade change? Pectra activated on **May 7th, 2025** and bundled **11 EIPs** across three user-facing goals: **more blob capacity for L2s** (EIP-7691 doubled the blob target from 3 to 6 per block, max 6→9), **smarter wallets** (EIP-7702 lets externally-owned accounts act as smart accounts — enabling gas sponsorship, paying fees in tokens other than ETH, and transaction batching via the new "Set Code" / Type 4 transaction type), and **staking upgrades** (EIP-7251 raised the effective validator cap from 32 ETH to 2,048 ETH so rewards can compound; EIP-7002 and EIP-6110 simplified the deposit and withdrawal pipeline). Live adoption tracker: [growthepie.com/quick-bites/pectra-upgrade](/quick-bites/pectra-upgrade). ### When did Pectra activate? Pectra activated on **May 7th, 2025** at slot 11,649,024 on the Ethereum mainnet beacon chain. The upgrade is the combined "Prague" execution-layer fork plus "Electra" consensus-layer fork — hence Pectra. ### How many EIPs did Pectra include? Eleven. The five most user-visible ones are EIP-7691 (more blobs), EIP-7702 (smarter wallets / Set Code transactions), EIP-7251 (raised validator cap to 2,048 ETH), EIP-7002 (execution-layer-triggered exits and withdrawals), and EIP-6110 (deposit pipeline simplification). The other six are protocol-internal improvements. See growthepie's [Pectra tracker](/quick-bites/pectra-upgrade) for live adoption data on the user-facing EIPs. ### What is EIP-7691 (more blobs)? EIP-7691 **doubled Ethereum's blob capacity** — the blob target per block went from **3 to 6**, and the maximum from **6 to 9**. Blobs are how L2 rollups post their transaction data to L1 for data availability. Doubling the target lowers fees for the same volume of L2 transactions and gives the blob-fee market more headroom before it kicks in via EIP-1559-style price increases. Rollups had been operating at full blob capacity for months before Pectra; EIP-7691 immediately relieved that pressure. ### How big was the impact on L2 fees? Pectra immediately roughly halved blob fees for L2s by doubling the capacity. The full effect was indirect: cheaper blobs meant L2s could pass savings through to users, and in the weeks after Pectra most major L2s saw their per-transaction fees drop noticeably. The relationship is approximate — blob fees only escalate via EIP-1559 once blob usage exceeds the target, so the *amount* of fee relief depended on demand at each moment. Growthepie's [Pectra tracker](/quick-bites/pectra-upgrade) has the live blob count vs target chart. ### What is EIP-7702 (smarter wallets)? EIP-7702 introduces a new transaction type — **"Set Code" or Type 4** — that lets a regular Ethereum wallet (an EOA — externally-owned account) temporarily act as a smart contract account for one transaction. This unlocks features that previously required users to switch to a dedicated smart-contract wallet: **gas sponsorship** (someone else pays the fee), **paying gas in tokens other than ETH** (e.g. USDC), **transaction batching** (multiple actions in one user-signed transaction), and **session keys / delegation** (a dApp signs on the user's behalf within preset limits). EIP-7702 doesn't require users to change wallets — existing EOAs can opt in. ### What is a "Set Code" or Type 4 transaction? A new Ethereum transaction format introduced by EIP-7702 that carries an authorization list: signed instructions telling the protocol to temporarily install bytecode on the sender's EOA. While the transaction executes, the EOA behaves like a smart contract account; after the transaction completes the original code (none, since it's an EOA) is restored. The Pectra tracker chart on growthepie counts Type 4 transactions per chain — that's the simplest live adoption signal for EIP-7702. ### Which apps benefit most from EIP-7702? Wallets and dApps that want a "no-friction" UX without forcing users to switch to smart-contract wallets. Day-one beneficiaries include account-abstraction-aware wallets, dApps doing gas sponsorship for new-user onboarding, and DeFi protocols that previously needed two transactions (approve + action) and can now batch them. The growthepie [Pectra tracker](/quick-bites/pectra-upgrade) charts per-chain Type 4 adoption (Ethereum L1, Base, OP Mainnet, Unichain, Arbitrum) so you can see which ecosystems are picking it up fastest. ### What changed for Ethereum stakers in Pectra? Three things, mostly via EIPs 7251, 7002, and 6110. **EIP-7251** raised the maximum effective stake per validator from 32 ETH to 2,048 ETH — large stakers can now consolidate many validators into one and let rewards compound rather than spinning up new validators every time the balance hits 32 ETH. **EIP-7002** lets validators trigger their own exits and partial withdrawals from the execution layer (no need to keep a separate withdrawal credential). **EIP-6110** moved the deposit pipeline onto the execution layer, removing the multi-hour deposit delay that previously existed. ### Why raise the validator cap to 2,048 ETH? Two reasons. First, **compounding** — at 32 ETH per validator, every reward over 32 ETH was idle until the operator manually deposited a new 32 ETH validator. Raising the cap to 2,048 ETH lets larger stakers (custodians, exchanges, big solo stakers) compound their rewards automatically. Second, **network-load reduction** — fewer validators means fewer messages on the consensus layer per epoch, which makes future consensus-layer upgrades easier. Small solo stakers can still run 32 ETH validators; the change is opt-in. ### How does Pectra compare to Dencun and Fusaka? **Dencun** (March 2024) introduced blobs (EIP-4844 / proto-danksharding) with a target of 3 blobs per block — the foundation of modern L2 scaling. **Pectra** (May 2025) doubled blob capacity to 6 and is the upgrade where Ethereum prioritised **user experience** for the first time (smart accounts via EIP-7702, easier staking). **Fusaka** (December 2025) added PeerDAS and tripled blob capacity again to a target of 14 by the time the BPO2 follow-up landed, plus introduced the EIP-7918 blob fee floor. So roughly: Dencun = blobs exist; Pectra = blobs scale 2× *and* UX improves; Fusaka = blobs scale ~5× and the economics tighten. ### Was Pectra the "user experience" upgrade? Yes — that's the framing. Past Ethereum upgrades have mostly been technical improvements (gas accounting, signature aggregation, new opcodes). Pectra was the first one where the user-visible features dominated the announcement: smart accounts (EIP-7702), cheaper L2 transactions (EIP-7691), and simpler staking (the EIP-7251 / 7002 / 6110 trio). The growthepie team called Pectra "less clicks, less signatures, more blobs" — which captures the change well. ### Did Pectra break anything for existing apps? No. Pectra was non-breaking — existing smart contracts, wallets, and L2 implementations continued to work without changes. EIP-7702 is opt-in (a wallet only acts as a smart account if the user signs an authorization); EIP-7691 affects validators and rollups but is transparent to user contracts; the staking EIPs only affect validators and stakers. Apps don't need to migrate. ### What changed for end users? A few things became possible (or noticeably cheaper) after Pectra: **wallets can offer gas sponsorship and tokenised gas payment** via EIP-7702 (no need to hold ETH for fees if the wallet supports it); **L2 transactions got cheaper** because blob capacity doubled; **staking became simpler** for users running their own validators. The user-facing changes are gated by wallet support — your wallet has to expose EIP-7702 features for you to use them. ### Where do these facts come from? Activation date and EIP numbers come from the [Ethereum Foundation's Pectra announcement](https://blog.ethereum.org/) and the [EIP repository](https://eips.ethereum.org/). User-facing summaries (blob target = 6, validator cap = 2,048 ETH, etc.) come from growthepie's [Pectra quick-bite](/quick-bites/pectra-upgrade) which tracks live adoption data. No editorial interpretation: every claim on this page is verifiable against the EIP text or the live tracker. ### How is Pectra tracked on growthepie? See the [Pectra tracker](/quick-bites/pectra-upgrade) for live charts: blob count vs target per Ethereum block (EIP-7691 adoption), and Set Code / Type 4 transaction count per chain (EIP-7702 adoption across Ethereum L1, Base, OP Mainnet, Unichain, Arbitrum). Both refresh daily from chain data. ### What came next after Pectra? Fusaka in December 2025 — see [/answers/what-fusaka-upgrade-changed](/answers/what-fusaka-upgrade-changed) for the full breakdown. Fusaka tripled blob capacity again (to a target of 14 by January 2026), introduced PeerDAS for safer blob scaling, and added EIP-7918 to put a floor under blob fees. After Fusaka comes Glamsterdam — see [ethereum.org/roadmap](https://ethereum.org/en/roadmap/) for the latest scheduled upgrades. --- ## Is Ethereum deflationary? URL: https://www.growthepie.com/answers/is-ethereum-deflationary ### Direct answer It depends on network activity. Ethereum has two opposing supply forces: EIP-1559 (active since August 2021) burns the base fee of every transaction, while Proof-of-Stake (since The Merge in September 2022) issues new ETH as staking rewards. When burn exceeds issuance, ETH is deflationary; when issuance exceeds burn, it is mildly inflationary. The Merge cut new ETH issuance by ~87%, so even modest network activity often pushes ETH into net-negative supply growth. The live annualised issuance rate and 30-day net change are tracked at growthepie.com/quick-bites/eth-supply. ### Full answer Short answer: Ethereum's annualised issuance rate is currently 0.85% (data live from growthepie's ETH supply tracker). Negative = deflationary (more ETH is being burned than created); positive = inflationary (more ETH is being created than burned). Net change in total supply over the last 30 days: Ξ81,431. Current total supply: Ξ121,728,733. Whether Ethereum is deflationary on any given day depends on network activity. High activity → more transactions → more base fees burned → deflationary. Low activity → less burn → inflationary. The live tracker is at growthepie.com/quick-bites/eth-supply. Two opposing forces Ethereum's supply is governed by two mechanisms working in opposite directions: 1. The Burn (EIP-1559, activated August 5th, 2021). Every transaction on Ethereum mainnet pays a base fee, calculated by the protocol based on how full the previous block was. That base fee is burned — permanently removed from circulation. Gas-heavy transactions (DeFi swaps, complex contracts) burn more ETH than simple transfers. The result: a direct link between network usage and ETH supply. 2. Issuance (Proof-of-Stake, since The Merge on September 15th, 2022). New ETH is created as staking rewards paid to validators in proportion to how much ETH is staked network-wide. The Merge cut total issuance by 87% — from roughly 13,000 ETH/day of mining rewards to 1,700 ETH/day of staking rewards. PoS issuance is now mathematically predictable and far smaller than pre-Merge. The interaction: when burn > issuance, ETH is deflationary. When issuance > burn, ETH is mildly inflationary. The post-Merge issuance floor is so low (0.5–1% per year) that even modest network activity often pushes ETH into deflationary territory. When has Ethereum been deflationary vs inflationary? - Pre-EIP-1559 (before August 2021) — purely inflationary. Mining issued 13,000 ETH/day with no burn. - Post-EIP-1559, pre-Merge (Aug 2021 → Sep 2022) — usually inflationary but with the burn beginning to offset issuance during high-activity periods. - Post-Merge (Sep 2022 → early 2024) — frequently deflationary. With issuance cut 87%, even moderate L1 activity often pushed net supply growth negative. - Post-Dencun (March 2024 onward) — closer to neutral, with stretches of mild inflation. Most routine activity moved to L2s, reducing L1 burn even though L2 settlement transactions still contribute via blob and calldata fees. - Post-Fusaka (December 2025) — EIP-7918's blob fee floor and the larger blob targets push burn higher than it would have been otherwise. See /answers/what-fusaka-upgrade-changed for the upgrade details. So "is Ethereum deflationary?" is best answered as: sometimes yes, sometimes no — depends on network activity in the window you measure. The current 30-day window gives you Ξ81,431 of net change. Why does Ethereum burn ETH? Three reasons, all from EIP-1559: - Predictable fees — the base fee is set by the protocol based on block fullness, not by user bidding wars. This makes fees more predictable. - MEV resistance — burning the base fee means validators can't collude with users to extract the entire fee, which would have made priority-fee bidding wars worse. - Value accrual — burning fees gives ETH holders a benefit from network activity even if they don't stake. The deflationary side effect is a consequence, not the goal of EIP-1559. Methodology and data sources All live numbers on this page come from growthepie's ETH supply tracker, which aggregates beacon chain data and burn events from Ethereum mainnet via . Updates daily. Specific placeholders: - 121,728,733 — total ETH supply at the latest day in the supply series. - 81,431 — . Negative = deflationary over the window. - 0.85 — latest value of the annualised issuance rate series, multiplied by 100. Negative = deflationary in annualised terms. Historical facts (EIP-1559 date, Merge date, 87% issuance reduction) come from published Ethereum Foundation announcements and the EIP repository. For per-block burn detail and a cumulative-burn counter, see ultrasound.money — that's the community-standard live resource. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. ETH supply data comes from canonical Ethereum mainnet sources — no supporter influences the reported numbers. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. Independent live burn / issuance trackers include ultrasound.money (per-block burn + cumulative since EIP-1559), Etherscan's supply page, and the ConsenSys ETH supply dashboard. Numbers can differ slightly between providers because of timing (different block-height snapshots) and source choice (CL vs EL data), but the direction and order of magnitude should always agree. ### FAQ ### Is Ethereum deflationary? **It depends on network activity.** ETH's supply is governed by two opposing forces — **issuance** (new ETH created as staking rewards) and **burn** (ETH destroyed by EIP-1559 base fees on every transaction). When burn exceeds issuance, ETH is deflationary; when issuance exceeds burn, it's inflationary. As of right now, Ethereum's **annualised issuance rate is %** — negative means deflationary, positive means inflationary. Net change in supply over the last 30 days: **Ξ** (negative = ETH was burned faster than created). Total supply: **Ξ**. Live tracker: [growthepie.com/quick-bites/eth-supply](/quick-bites/eth-supply). ### How does ETH get burned? **EIP-1559** ("The Burn", activated August 5th, 2021) split every transaction fee into two parts: a **base fee** that gets burned (destroyed forever) and a **priority fee / tip** that goes to the validator who included the transaction. The base fee algorithmically adjusts based on block fullness — busier blocks have a higher base fee, so more ETH burns. This means **every transaction on Ethereum mainnet permanently removes a small amount of ETH from circulation** — gas-heavy transactions (DeFi swaps, complex contract interactions) burn more ETH than simple transfers. ### How does new ETH get created (issuance)? After **The Merge** (September 15th, 2022), Ethereum transitioned from Proof-of-Work to Proof-of-Stake. Mining stopped — and with it, ~13,000 ETH/day of mining rewards. Staking rewards (~1,700 ETH/day at the time) replaced them, **cutting new ETH issuance by roughly 87% overnight**. New ETH now enters circulation as staking rewards paid to validators in proportion to how much ETH is staked network-wide. ### When is Ethereum deflationary? When **burn > issuance** — typically during periods of high network activity. Historical examples: most of late 2022 and 2023 (post-Merge but pre-Dencun) saw extended deflationary stretches because L1 activity stayed high and burned more ETH than staking issued. Periods of low L1 activity (post-Dencun in 2024, as routine transactions moved to L2s) tend to be slightly inflationary because issuance is now mostly higher than the residual L1 burn. The answer to "is ETH deflationary today" therefore varies day-to-day — the current annualised rate is **%**. ### When is Ethereum inflationary? When **issuance > burn** — typically during periods of low network activity. When L1 transactions are cheap (e.g. low demand, or after upgrades like Dencun moved most routine activity to L2s) the base fee stays low, less ETH burns, and staking issuance dominates. The annualised inflation rate stays low even at peak inflation (the post-Merge issuance floor is roughly 0.5–1% per year), so even "inflationary Ethereum" is a far slower-issuing asset than pre-Merge ETH was. ### How much ETH has been burned via EIP-1559? Cumulatively, several million ETH have been burned since EIP-1559 activated in August 2021. The exact total updates with every block — see [ultrasound.money](https://ultrasound.money/) for the live cumulative burn counter, and growthepie's [ETH supply tracker](/quick-bites/eth-supply) for the supply curve over time (the visible "kink" at The Merge is where issuance dropped 87%; the slower curve after EIP-1559 is the burn at work). ### Did The Merge make Ethereum deflationary? Not by itself — **The Merge cut issuance, EIP-1559 added the burn**. The two together create the conditions for net-negative supply growth. The Merge alone (September 2022) reduced new ETH issuance by ~87% but didn't introduce any burn mechanism — that was already in place from EIP-1559 (August 2021). The combination is what makes ETH "ultrasound money": even modest network activity now burns enough ETH to offset the (much smaller) staking issuance. ### What is "ultrasound money"? A community term for an asset whose supply *decreases* over time — going one step further than "sound money" (a fixed supply asset like Bitcoin). The phrase was coined to describe Ethereum's post-Merge potential to become deflationary via EIP-1559 burn outpacing PoS issuance. Whether ETH actually is "ultrasound money" depends on which period you measure — see the historical breakdown in this FAQ. ### Why does Ethereum burn ETH at all? Three reasons, all from EIP-1559: **(1) Predictable transaction fees** — the base fee is set by the protocol based on block fullness, not by users bidding against each other. This makes fees more predictable. **(2) MEV resistance** — burning the base fee means validators can't collude with users to extract the entire fee, which would have made priority fee bidding wars worse. **(3) Value accrual** — burning fees gives ETH holders a benefit from network activity even if they don't stake. The deflationary side effect is a consequence, not the goal. ### How is the burn rate calculated? For each block, the burn equals `gas_used × base_fee_per_gas`. The base fee adjusts each block based on the previous block's fullness — if blocks are above 50% full, the base fee increases; below 50% full, it decreases. So **high network activity → fuller blocks → higher base fee → more ETH burned per transaction**. Tools like growthepie's [ETH supply tracker](/quick-bites/eth-supply) and [ultrasound.money](https://ultrasound.money/) aggregate this per-block burn into daily / weekly / annualised rates. ### Do L2 transactions burn ETH? **Indirectly, yes.** Every L2 rollup posts its data to Ethereum mainnet as either calldata (pre-Dencun) or blobs (post-Dencun, March 2024). That L1 transaction *does* pay a base fee, which gets burned. So even though L2 users don't directly burn ETH for each L2 transaction, the L2's settlement transaction on L1 burns ETH on the user's behalf. After Dencun the burn-per-L2-transaction dropped sharply (blobs are way cheaper than calldata), which is one reason ETH net issuance ticked up from late 2024 onward. ### Does the Fusaka upgrade affect the burn? **Yes — in two ways.** First, EIP-7918 (the blob fee floor) ended the "zero-fee blob era" and ensures blob fees track L1 execution prices, which contributes to ongoing burn. Second, the larger blob target raised by Fusaka and the BPO upgrades means more blob fees collected per block during high L2 demand. Both push burn higher than it would have been without Fusaka. See [/answers/what-fusaka-upgrade-changed](/answers/what-fusaka-upgrade-changed) for the full breakdown. ### Where can I see live burn data? growthepie's [ETH supply tracker](/quick-bites/eth-supply) charts total supply and issuance rate over time with The Merge and EIP-1559 marked as annotations. For per-block burn detail and the exact cumulative burn counter, [ultrasound.money](https://ultrasound.money/) is the community-standard resource. ### How does Ethereum's monetary policy compare to Bitcoin's? Bitcoin has a **fixed total supply** (21M BTC, with new issuance halving every ~4 years) and **no burn mechanism**. ETH has **no fixed supply cap** but the burn mechanism means total supply can decrease in periods of high activity — something Bitcoin's design can't do. Practically: Bitcoin's supply growth is mechanically predictable, Ethereum's depends on network usage. Both are far more constrained than fiat issuance, but they're structurally different. ### Where does this answer come from? Live numbers (total supply, 30-day net issuance, annualised issuance rate) come from growthepie's ETH supply endpoint (`/v1/eim/eth_supply.json`), which aggregates beacon chain data and burn events from Ethereum mainnet. The endpoint is the same one that backs growthepie's [ETH supply tracker](/quick-bites/eth-supply). Historical context (EIP-1559 activation, The Merge, percentage issuance reduction) comes from the published EIPs and the Ethereum Foundation's announcements. No editorial interpretation: every claim here is verifiable against the tracker or the source EIPs. --- ## What is the difference between a ZK rollup and an Optimistic rollup? URL: https://www.growthepie.com/answers/zk-vs-optimistic-rollup ### Direct answer ZK rollups and Optimistic rollups are both Ethereum Layer 2s that batch transactions and post data to mainnet. They differ in how they prove batches are correct: ZK rollups generate cryptographic validity proofs verified by Ethereum (fast L1 finality, minutes to hours); Optimistic rollups assume batches are valid and rely on a 7-day fraud-proof challenge window. Both inherit Ethereum's security. Major ZK rollups: zkSync Era, Polygon zkEVM, Linea, Scroll, Starknet, Taiko. Major Optimistic rollups: Arbitrum One, OP Mainnet, Base, Mantle, Worldchain, Unichain, Soneium. Most newer L2s have chosen the ZK path. ### Full answer Short answer: ZK rollups prove batches are valid with a cryptographic proof verified by Ethereum (fast finality, mathematical security). Optimistic rollups assume batches are valid and rely on a 7-day fraud-proof window (simpler engineering, slower L1 finality). Both inherit Ethereum's security; the difference is the proof system. Both are Ethereum Layer 2s. They batch transactions, post the data to Ethereum, and inherit Ethereum's security. The only architectural difference is how they convince Ethereum the batches are correct. The core difference Imagine you're processing transactions in batches off-chain and need Ethereum to trust the result. ZK rollup: "Here's the batch, and here's a cryptographic proof that every transaction in it was valid. Ethereum, verify the proof." If the proof verifies, the batch is final. Optimistic rollup: "Here's the batch. Assume it's correct unless someone proves otherwise within 7 days." Anyone can submit a fraud proof during the window; if it verifies, the bad batch is reverted. Same security guarantee in the limit (both inherit Ethereum), but the mechanics — and the user experience — differ. Which Ethereum L2s are which? As of 2026, the major chains break down roughly like this: ZK rollups — zkSync Era, Polygon zkEVM, Linea, Scroll, Starknet (non-EVM), Taiko (based ZK rollup), and a growing list of newer chains. Nearly every new L2 announced from 2024 onward has chosen the ZK path. Optimistic rollups — Arbitrum One, OP Mainnet, Base, Mode, Mantle, Worldchain, Unichain, Soneium, Ink, Lisk, and many others. Most are built on either the OP Stack (Optimism's open-source codebase) or Arbitrum Orbit (Arbitrum's codebase). See growthepie.com/chains for the live list with each chain's current classification. Trade-offs A practical comparison of the trade-offs that actually matter to users and developers: | Aspect | ZK rollup | Optimistic rollup | |---|---|---| | L1 finality | Minutes–hours | 7 days (challenge window) | | Native withdrawal time | Minutes–hours | 7 days (or via 3rd-party fast bridge) | | Security model | Cryptographic — proof must verify | Cryptoeconomic — at least one honest verifier | | EVM compatibility | Varies (some non-EVM, most EVM-compatible) | Almost always EVM-equivalent | | Engineering complexity | High (ZK circuits are hard) | Lower (fraud proofs reuse EVM) | | Proof costs | Pays for every proof | Only pays during a fraud challenge | | Live examples | zkSync, Linea, Scroll, Starknet, Polygon zkEVM | Arbitrum, OP, Base, Mantle, Worldchain | In practice, users feel the withdrawal time difference most (Optimistic chains require 7-day withdrawals to L1 unless you use a fast bridge), and developers feel the EVM compatibility difference (Optimistic chains tend to be plug-and-play; some ZK chains require contract adjustments). Which is more secure? Both inherit Ethereum's security in principle, but the practical model differs: - ZK rollups have mathematical security — if the proof verifies, the state transition is correct. No humans in the loop, no liveness assumptions about verifiers. - Optimistic rollups have cryptoeconomic security — they require at least one honest verifier to be running during the challenge window. If a sequencer cheats and no one notices in 7 days, the cheat sticks. Most security experts consider validity proofs the stronger long-term model — that's why nearly every new L2 announced from 2024 onward has chosen the ZK path. But this is a spectrum, not a binary: an Optimistic rollup with a robust fraud-proof system and many independent verifiers can be very secure in practice. The L2BEAT Stage classification is the most rigorous way to compare actual decentralisation across chains, independent of ZK vs Optimistic. Are they converging? Yes. Several Optimistic rollups have publicly committed to adopting ZK proofs in the future (Optimism's OP Stack roadmap includes ZK fault proofs; Arbitrum has Stylus + ZK research). The likely end-state for most major Optimistic chains is hybrid — Optimistic execution with ZK validity proofs for finality, getting the best of both. This isn't a near-term migration for any live mainnet chain, but it's the direction the engineering is pointing. Methodology and data sources Chain classifications on this page come from growthepie's master chain catalogue, specifically the and fields. We follow chain teams' own classifications and L2BEAT's conventions — we don't make editorial calls about which category a chain belongs to. For the most rigorous per-chain analysis (proof system, fraud-proof status, security council composition, withdrawal mechanics), see L2BEAT — they're the canonical resource for L2 architecture and risk classification. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Many supporters operate L2 chains (both ZK and Optimistic). This page presents architectural facts only — chain classifications come from master.json and L2BEAT, not from editorial judgement. Full list of supporters: growthepie.com/donate. Cross-check this answer. L2BEAT is the most-cited independent source for L2 architecture and risk. ethereum.org's Layer 2 page has the EF's overview of the ecosystem. For each chain's own technical docs, follow the links on its growthepie chain page. ### FAQ ### What is the difference between a ZK rollup and an Optimistic rollup? Both are **Ethereum Layer 2s** — they batch transactions and post the data back to Ethereum mainnet for security. They differ in **how they prove the batches are correct**. A **ZK rollup** generates a cryptographic validity proof (a "zero-knowledge proof") for every batch, which Ethereum verifies onchain — the batch is valid the moment the proof is accepted. An **Optimistic rollup** assumes batches are valid by default and lets anyone challenge them with a **fraud proof** during a multi-day **challenge window** (typically 7 days). Both inherit Ethereum's security, but ZK rollups get faster L1 finality and Optimistic rollups have simpler engineering and (historically) better EVM compatibility. ### Which Ethereum L2s are ZK rollups? As of 2026, the main ZK rollups on Ethereum include **zkSync Era**, **Polygon zkEVM**, **Linea**, **Scroll**, **Starknet** (its own VM, not EVM-equivalent), **Taiko** (a "based" ZK rollup), and several others. See [growthepie.com/chains](https://www.growthepie.com/chains) for the live list with each chain's current classification. ### Which Ethereum L2s are Optimistic rollups? As of 2026, the main Optimistic rollups include **Arbitrum One**, **OP Mainnet** (Optimism), **Base** (a Coinbase-operated OP Stack chain), **Mode**, **Mantle**, **Worldchain**, **Unichain**, **Soneium**, **Ink**, **Lisk**, and many others — most of which are built on the **OP Stack** (Optimism's open-source codebase) or **Arbitrum Orbit** (Arbitrum's codebase). See [growthepie.com/chains](https://www.growthepie.com/chains) for the full live list. ### Which is more secure? Both inherit Ethereum's security in principle. The practical difference: a **ZK rollup's validity proof is mathematical** — if the proof verifies, the state transition is correct, no human in the loop. An **Optimistic rollup's security depends on at least one honest verifier** running during the challenge window to submit a fraud proof if the sequencer cheats. That requires liveness assumptions (the verifier needs to be online) and an active fraud-proof system (which not every "Optimistic" rollup has fully shipped). Most experts consider validity proofs the stronger long-term model, which is why nearly every new L2 announced from 2024 onward has chosen the ZK path. ### Which is faster? For **L1 finality** (when the L2's state is settled on Ethereum), **ZK rollups are faster** — typically minutes to hours from when a batch is posted to when its proof is verified. Optimistic rollups have a **7-day challenge window** before L1 considers the batch final, which is also when withdrawals can be processed. For **user experience** (UX finality on the L2 itself — what your wallet shows), both are effectively instant — usually under a second. Most users never notice the L1 finality difference because L2-native applications treat L2 confirmations as final. ### Why do Optimistic rollups need a 7-day challenge window? To give honest verifiers enough time to detect and challenge fraud. If a sequencer posts an invalid state, anyone can submit a fraud proof during the window; if the fraud proof verifies, the bad batch is reverted. The window has to be long enough that a censorship attack (where an attacker tries to suppress fraud proofs) can't succeed. **Bridges built on top of Optimistic rollups often abstract this away** — they let users withdraw quickly via third-party liquidity providers who front the ETH/USDC and wait the 7 days themselves, charging a small fee for the service. ### Why have most newer L2s chosen ZK? Three reasons. **(1) Better finality** — no 7-day challenge window means simpler bridges, faster withdrawals, and cleaner UX. **(2) Cleaner security model** — validity proofs don't require trust assumptions about verifier liveness. **(3) Hardware improvements** — proving zkEVM circuits used to be slow and expensive; advances in ZK proving (and dedicated proving hardware) have made it economically viable. The trade-off is engineering complexity: ZK proof systems are much harder to design and audit than fraud-proof systems. ### Are ZK rollups always EVM-compatible? Not always. The spectrum runs from **fully EVM-equivalent** (your Ethereum contract works as-is — Linea, Scroll, Polygon zkEVM aim for this) to **EVM-compatible with caveats** (most contracts work but some opcodes behave differently — early zkSync Era was here) to **non-EVM** (Starknet uses its own VM called Cairo). Optimistic rollups are almost always **EVM-equivalent** by construction because their fraud-proof system runs the EVM directly. The compatibility gap is closing fast on the ZK side as proving tech matures. ### Which has cheaper fees? Per-transaction L2 fees depend more on **the chain's blob/calldata strategy** and **the chain's current activity level** than on whether it's ZK or Optimistic. Both ZK and Optimistic rollups post their batch data to Ethereum as blobs (post-Dencun), and that DA cost is by far the largest fee component for either. ZK rollups also have **proof generation cost** which is non-trivial; Optimistic rollups don't pay for proofs unless a fraud claim is made. For live per-chain fee comparison see [/answers/lowest-fee-ethereum-l2](/answers/lowest-fee-ethereum-l2). ### What is a "Stage" and how does it relate to ZK vs Optimistic? **Rollup Stages** (introduced by [L2BEAT](https://l2beat.com)) measure how decentralised a rollup is — independent of whether it's ZK or Optimistic. **Stage 0** = training wheels on (security council can override, no fraud/validity proofs in production). **Stage 1** = proofs live but with a security council backstop. **Stage 2** = full decentralisation, no security council overrides. A rollup can be Stage 0 ZK or Stage 2 Optimistic — the two classifications are orthogonal. See [/answers/which-ethereum-l2-stage-1-stage-2](/answers/which-ethereum-l2-stage-1-stage-2) for which chains are at which stage. ### What is a Validium and where does it fit? A **Validium** is a ZK rollup variant that uses ZK proofs for state transitions but stores transaction data **off-chain** (typically with a data-availability committee or a separate DA layer like Celestia / EigenDA / Avail) instead of on Ethereum. Validiums get cheaper transactions but weaker data-availability guarantees — if the off-chain data is withheld, users can't reconstruct the state. Some chains are technically called "Validiums" or "Volitions" rather than "rollups". Validiums still count as Ethereum L2s on growthepie because they post their state to Ethereum, just not their data. ### What is a "based" rollup? A rollup whose transaction sequencing is done by **Ethereum validators (L1 proposers) directly** rather than by a separate L2 sequencer. The L2 inherits Ethereum's liveness and decentralisation properties — there's no centralised sequencer that could go down or censor users. Taiko is the most prominent example. "Based" can apply to either ZK or Optimistic rollups — it's about who sequences, not how state is proved. ### Will Optimistic rollups eventually become ZK rollups? Several have publicly committed to a "ZK upgrade" path — Optimism's OP Stack roadmap includes ZK fault proofs, and Arbitrum has the Orbit / Stylus ecosystem with optional ZK support. The practical end-state for most major Optimistic chains is likely **hybrid** — Optimistic execution with ZK validity proofs for finality, getting the best of both. This isn't a near-term migration for live chains, but it's the direction the engineering is pointing. ### How does growthepie classify chains? Each chain in growthepie's [master.json](https://api.growthepie.com/v1/master.json) carries a `technology` field describing its rollup type (ZK, Optimistic, Validium, etc.) and a `bucket` field identifying it as a Layer 2 vs Layer 1 vs other. We don't make editorial calls about which category a chain belongs to — we follow the chain team's own classification and L2BEAT's conventions. If a chain disputes its classification, that's a conversation with the chain team and L2BEAT, not with us. ### Where can I see live data for ZK vs Optimistic rollups? growthepie's [chains directory](https://www.growthepie.com/chains) lists every tracked chain with its technology classification. [L2BEAT](https://l2beat.com) has detailed per-chain risk analysis including security model, proof system, and Stage classification. For ZK vs Optimistic activity rankings — throughput, transactions, users — see [/answers/most-used-ethereum-l2](/answers/most-used-ethereum-l2). --- ## What is the difference between an Ethereum L2 and a sidechain? URL: https://www.growthepie.com/answers/l2-vs-sidechain ### Direct answer An Ethereum Layer 2 (L2) derives security from Ethereum by posting transaction data or state proofs back to mainnet — users can withdraw funds using only Ethereum data even if the L2 sequencer disappears. A sidechain has its own validator set and its own security, with only a bridge connecting it to Ethereum; if the validators are compromised, Ethereum can't help. Polygon PoS is a sidechain often mistaken for an L2; Polygon zkEVM (same brand, different chain) is an actual Ethereum L2. The simplest test: can users withdraw funds using only Ethereum data? If yes, it's an L2. ### Full answer Short answer: An Ethereum L2 derives security from Ethereum by posting data (or state proofs) back to mainnet — users can withdraw funds using only Ethereum data even if the L2 sequencer disappears. A sidechain has its own validator set and its own security; the only thing connecting it to Ethereum is a bridge. Polygon PoS is the most famous sidechain often mistaken for an L2 (Polygon zkEVM is the actual Polygon L2 — same brand, completely different architecture). The simplest test: "if this chain went away tomorrow, could users withdraw their funds using only Ethereum?" If yes, it's an L2. If no, it's a sidechain. The core distinction Ethereum L2s post their transaction data (or cryptographic proofs of state) back to Ethereum mainnet. This means Ethereum holds an authoritative copy of the L2's state — if the L2 sequencer disappears or cheats, anyone can use the Ethereum data to reconstruct user balances and withdraw funds. The security model: L2 inherits Ethereum. Sidechains run as separate blockchains with their own validator sets and their own consensus. They have a bridge to Ethereum, but the bridge only trusts the sidechain's validators — Ethereum doesn't verify the sidechain's state. If the validator set is compromised, the bridge can be drained. The security model: sidechain has its own. The 2022 Ronin bridge hack ($600M stolen) is the textbook example of how this can go wrong: an attacker compromised the Ronin validator set (Ronin was a sidechain at the time — it has since migrated to an Ethereum L2 architecture) and used the bridge's "trust the validators" model to drain the bridge contract. Examples Ethereum L2s: Arbitrum One, OP Mainnet, Base, zkSync Era, Polygon zkEVM, Linea, Scroll, Mantle, Worldchain, Unichain, Soneium, and many more. All post data to Ethereum. Sidechains: Polygon PoS (100 validators, own consensus), Gnosis Chain (formerly xDai), BNB Smart Chain (Binance's EVM chain — often called a "BNB sidechain" but technically a separate alt-L1). None of these post their state to Ethereum. Former sidechain, now an L2: Ronin (Axie Infinity's chain) was a sidechain through 2024 and is the source of the most-cited sidechain bridge hack — the 2022 $600M Ronin Bridge compromise. It has since migrated to an Ethereum L2 architecture and is no longer classified as a sidechain. Note: Polygon zkEVM and Polygon PoS share a brand but are completely different architectures. zkEVM is an Ethereum L2 (ZK rollup); PoS is a sidechain. Why this matters The practical consequence is what could go wrong: - On an Ethereum L2, the worst plausible failure is the sequencer goes offline. Funds are still recoverable from Ethereum. - On a sidechain, the worst plausible failure is the validator set is compromised (collusion, hack, regulatory seizure). Funds are gone. For high-value use cases (DeFi, treasury management, NFTs with real value) this gap matters. For low-value gaming, social applications, or experiments, the sidechain trade-off may be acceptable for the lower fees and higher throughput. In 2026 the sidechain speed/cost argument is also weaker than it used to be — major Ethereum L2s now have throughput and fees competitive with sidechains, with stronger security guarantees. Where Validiums and Volitions sit Validium = ZK rollup-like state proofs, sidechain-like off-chain data. Volition = user picks per-transaction whether data goes on-chain or off-chain. Both are usually classified as L2s on growthepie because the state is provably secured by Ethereum — but they're a partial step from "true rollup" toward "sidechain-like data assumptions". L2BEAT classifies them as "Validium" with explicit risk caveats. The category line isn't binary, but it's sharper than people often assume. Methodology and data sources Chain classifications on this page come from growthepie's master chain catalogue, specifically the field. We use the same field to filter the L2 universe on every other answer page on growthepie — so chains classified as sidechains (Polygon PoS, BNB Smart Chain, Gnosis Chain, etc.) are deliberately excluded from L2 rankings. When a chain migrates from sidechain to L2 (as Ronin did), growthepie updates the bucket classification in master.json and the chain begins appearing in L2 rankings automatically. We follow L2BEAT's classification conventions — they're the canonical independent resource for L2 architecture. If a chain disputes its classification, that's a conversation with L2BEAT and the chain team, not editorial judgement on our side. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Several supporters are L2 chains. Sidechain classifications on this page follow L2BEAT and master.json — no editorial bias toward or against any chain. Full list of supporters: growthepie.com/donate. Cross-check this answer. L2BEAT is the most-cited independent classifier of L2s vs sidechains — they explicitly publish a stage and bucket for every project. ethereum.org's Layer 2 page has the EF's overview of what constitutes an L2. Both should agree with growthepie's classification. ### FAQ ### What is the difference between an Ethereum L2 and a sidechain? An **Ethereum Layer 2 (L2)** is a chain that **derives its security from Ethereum** by posting transaction data (or a cryptographic proof of state) back to Ethereum mainnet. If the L2's sequencer disappears or cheats, users can recover their funds using only Ethereum data. A **sidechain** is a separate blockchain with its **own validator set, its own security**, and its own bridge to Ethereum — it doesn't post anything to Ethereum for security purposes. Polygon PoS is the most famous sidechain often confused with an L2; BNB Smart Chain and Gnosis Chain are other examples. (Note: Ronin was a sidechain through 2024 — including the 2022 bridge hack — but has since migrated to an L2 architecture.) L2s share Ethereum's trust assumptions; sidechains require trusting their own validators. ### Is Polygon PoS an Ethereum L2? **No** — Polygon PoS is a **sidechain**, not an L2. It has its own ~100-validator Proof-of-Stake set, runs its own consensus, and does not post transaction data to Ethereum for security. The bridge between Polygon PoS and Ethereum is just a bridge — if the Polygon validator set were compromised, Ethereum couldn't help users recover their funds. **Polygon zkEVM**, on the other hand, *is* an Ethereum L2 (specifically a ZK rollup) — same brand, completely different architecture. L2BEAT also classifies Polygon PoS as "other" rather than as a Layer 2. ### Why does this distinction matter? **Security guarantees.** When you bridge $100 of USDC to an Ethereum L2 like Arbitrum or Base, the worst case is the sequencer goes offline — you can still recover your funds because Ethereum has all the data. When you bridge $100 of USDC to a sidechain like Polygon PoS, the worst case is the validator set is compromised — at which point your funds are gone, because Ethereum has no record of the sidechain's state. The trust assumption is fundamentally different. For high-value use cases (DeFi, settlement), this matters; for low-value gaming or social, the sidechain trade-off may be acceptable. ### How do you tell an L2 from a sidechain at a glance? **Ask: "if this chain went away tomorrow, could users withdraw their funds using only Ethereum?"** If yes, it's an L2 — the data needed to reconstruct user balances lives on Ethereum (either as calldata, blobs, or compressed state). If no, it's a sidechain — the chain's own validator set holds the only authoritative copy of the state. **L2BEAT** is the canonical resource for this classification; growthepie follows L2BEAT's conventions in our [master chain catalogue](https://api.growthepie.com/v1/master.json). ### What are the most well-known Ethereum sidechains? **Polygon PoS** (formerly Matic) — the most prominent. ~100 validators, its own PoS consensus, two-way bridge to Ethereum. **Gnosis Chain** (formerly xDai) — a DAI-stablecoin-native chain with its own consensus. **BNB Smart Chain** — Binance's EVM-compatible chain, often confused with an Ethereum L2 (it isn't). **Ronin** was a sidechain through 2024 (and is the source of the most-cited sidechain bridge hack, the $600M Ronin Bridge compromise in 2022) but has since **migrated to an Ethereum L2** — it's no longer a sidechain. Several other former sidechains are on similar paths: Polygon's "Polygon 2.0" plan involves shifting Polygon PoS toward a validium architecture too. ### What are the most well-known Ethereum L2s? By usage today: **Base** (Coinbase's L2 on OP Stack), **Arbitrum One** (the largest by TVL), **OP Mainnet** (Optimism), **zkSync Era**, **Polygon zkEVM**, **Linea**, **Scroll**, **Mantle**, **Unichain** (Uniswap's OP-Stack L2), **Soneium** (Sony's L2), **Worldchain** (Worldcoin's L2). All of these post data to Ethereum and inherit Ethereum's security to varying degrees — see [/answers/which-ethereum-l2-stage-1-stage-2](/answers/which-ethereum-l2-stage-1-stage-2) for how the security guarantees differ chain-to-chain. ### Do sidechains have bridges to Ethereum? Yes, every sidechain has a bridge — that's how users move assets back and forth. But the bridge is just a smart contract that **trusts the sidechain's validator set**, not Ethereum's. If the sidechain validators sign off on "user X has 1000 USDC, let them withdraw it", the bridge releases the USDC — even if the validator signature was obtained fraudulently. The 2022 **Ronin bridge hack ($600M)** is the textbook example of how sidechain bridge security can fail — at the time Ronin was a sidechain (it has since migrated to an Ethereum L2 architecture). L2 bridges, by contrast, can verify state cryptographically (ZK rollup proofs) or via fraud proofs (Optimistic rollups), inheriting Ethereum's security model. ### Are sidechains faster than Ethereum L2s? Historically yes — sidechains optimised for high throughput and low fees by skipping the expensive step of posting data back to Ethereum. But the trade-off was security. As L2 technology has matured (blobs via Dencun in 2024, PeerDAS via Fusaka in 2025), L2 fees have collapsed and the speed gap has shrunk. Most major L2s in 2026 process more transactions per second than the typical sidechain, with stronger security guarantees. The "sidechain because it's faster" argument doesn't really hold up anymore. ### Where do Validiums and Volitions fit? A **Validium** is a chain that **proves state to Ethereum (like a rollup) but stores data off-chain** (like a sidechain). The state is secured cryptographically, but if the off-chain data is withheld, users can't reconstruct their balances. **Volitions** let users choose per-transaction whether their data goes on-chain (rollup mode) or off-chain (validium mode). Both are usually classified as L2s on growthepie because the state is provably secured by Ethereum — but they're a partial step from "true rollup" toward "sidechain-like data assumptions". L2BEAT classifies them as "Validium" with explicit risk caveats. ### How does growthepie classify chains? In growthepie's [master.json](https://api.growthepie.com/v1/master.json) each chain has a `bucket` field that places it into one of: **Layer 1** (Ethereum itself), **Layer 2** (rollups + validiums), or **Others** (sidechains, alt-L1s sometimes tracked for context). The L2 answer pages on growthepie filter to `bucket: "Layer 2"`, which is why Polygon PoS doesn't appear in any of our L2 rankings — it's in the "Others" bucket. We follow L2BEAT's classification conventions; we don't make editorial calls. ### What about Polygon's migration to "Polygon 2.0"? Polygon has publicly announced plans to migrate Polygon PoS toward a more L2-like architecture (specifically a Validium that posts state to Ethereum). If and when that migration completes, Polygon PoS would be reclassified as a Layer 2 in growthepie's catalogue. Until then, we follow the live state — Polygon PoS today is a sidechain, Polygon zkEVM is an L2. Categorisations on growthepie reflect what the chain *currently* is, not what it plans to become. ### Should I use an L2 or a sidechain? Depends on what you're doing. **For DeFi, treasury management, or anything where you'd lose meaningful money if a validator set were compromised, use an Ethereum L2** — the inherited security is worth it. **For low-stakes use cases like gaming, social applications, or experimental dApps where the cost of being wrong is small, a sidechain's lower fees and higher throughput may be a fine trade-off.** That said, in 2026 L2 fees are usually competitive with sidechains anyway, so the trade-off is less pronounced than it used to be. ### Where can I see live data for L2s vs sidechains? growthepie's [chains directory](https://www.growthepie.com/chains) lists every tracked chain with its bucket (Layer 1 / Layer 2 / Others). [L2BEAT](https://l2beat.com) has detailed per-chain risk analysis and is the canonical source for the "is this really an L2?" question. For per-chain activity rankings restricted to actual L2s, see [/answers/most-used-ethereum-l2](/answers/most-used-ethereum-l2) — sidechains are deliberately excluded. --- ## Which Ethereum L2 has the most total value secured (TVS)? URL: https://www.growthepie.com/answers/most-value-secured-ethereum-l2 ### Direct answer By total value secured (TVS), the largest Ethereum L2 is **Arbitrum One** at $15.87B — 46% of the $34.56B L2 ecosystem total. Top 10: 1. Arbitrum One ($15.87B, 46% of L2 TVS; +1.9% 30d); 2. Base Chain ($12.15B, 35% of L2 TVS; +1.7% 30d); 3. OP Mainnet ($1.53B, 4.4% of L2 TVS; -2.8% 30d); 4. Mantle ($1.44B, 4.2% of L2 TVS; -11.1% 30d); 5. Starknet ($625.74M, 1.8% of L2 TVS; +17.8% 30d); 6. Linea ($447.59M, 1.3% of L2 TVS; -9.1% 30d); 7. Ink ($413.63M, 1.2% of L2 TVS; -3.2% 30d); 8. World Chain ($330.09M, 1.0% of L2 TVS; +19.5% 30d); 9. ZKsync Era ($292.37M, 0.8% of L2 TVS; -11.1% 30d); 10. Celo ($271.19M, 0.8% of L2 TVS; +2.0% 30d). Data: 2026-05-27 UTC. Live leaderboards: growthepie.com/chains. ### Full answer Short answer (data 2026-05-27 UTC): Arbitrum One ($15.87B, 46% of L2 TVS; +1.9% 30d) is the Ethereum L2 with the most total value secured. The combined L2 ecosystem total is approximately $34.56B across 25 tracked chains. Full top 10 by TVS: 1. Arbitrum One ($15.87B, 46% of L2 TVS; +1.9% 30d); 2. Base Chain ($12.15B, 35% of L2 TVS; +1.7% 30d); 3. OP Mainnet ($1.53B, 4.4% of L2 TVS; -2.8% 30d); 4. Mantle ($1.44B, 4.2% of L2 TVS; -11.1% 30d); 5. Starknet ($625.74M, 1.8% of L2 TVS; +17.8% 30d); 6. Linea ($447.59M, 1.3% of L2 TVS; -9.1% 30d); 7. Ink ($413.63M, 1.2% of L2 TVS; -3.2% 30d); 8. World Chain ($330.09M, 1.0% of L2 TVS; +19.5% 30d); 9. ZKsync Era ($292.37M, 0.8% of L2 TVS; -11.1% 30d); 10. Celo ($271.19M, 0.8% of L2 TVS; +2.0% 30d). Updated daily — every value on this page is recomputed from growthepie's per-chain TVL endpoint. TVS = total value secured = dollar value of all assets bridged to or natively held on the L2. Top 10 Ethereum L2s by TVS As of 2026-05-27 UTC, Arbitrum One leads Ethereum L2s by total value secured at $15.87B (46% of the L2 ecosystem total $34.56B). Top 10: 1. Arbitrum One ($15.87B, 46% of L2 TVS; +1.9% 30d); 2. Base Chain ($12.15B, 35% of L2 TVS; +1.7% 30d); 3. OP Mainnet ($1.53B, 4.4% of L2 TVS; -2.8% 30d); 4. Mantle ($1.44B, 4.2% of L2 TVS; -11.1% 30d); 5. Starknet ($625.74M, 1.8% of L2 TVS; +17.8% 30d); 6. Linea ($447.59M, 1.3% of L2 TVS; -9.1% 30d); 7. Ink ($413.63M, 1.2% of L2 TVS; -3.2% 30d); 8. World Chain ($330.09M, 1.0% of L2 TVS; +19.5% 30d); 9. ZKsync Era ($292.37M, 0.8% of L2 TVS; -11.1% 30d); 10. Celo ($271.19M, 0.8% of L2 TVS; +2.0% 30d). What does "TVS" mean exactly? Total Value Secured is the total USD value of all assets the chain holds on behalf of its users — stablecoins (USDC, USDT, DAI, PYUSD), bridged ETH, native ETH, ERC-20 tokens, wrapped Bitcoin, liquid staking tokens, NFTs, and any other on-chain asset. It's the standard "how much capital lives here" metric for L2s. You'll also see this metric called TVL (Total Value Locked) — that's the older term, originally from DeFi protocols where "locked" meant deposited into smart contracts. TVS is the newer label preferred for L2s because the chain secures the assets (users can withdraw them back to Ethereum) rather than locking them in a single application. The underlying number is the same; growthepie's API uses the endpoint name for both labels. TVS vs other "size" metrics TVS measures capital weight — how much value is parked on the chain. It's different from: - Throughput / transactions (/answers/most-used-ethereum-l2) — measures flow, not stock. A chain can have low TVS but high activity (lots of micro-transactions, cheap fees) or vice versa. - Stablecoin supply (/answers/most-stablecoin-activity-ethereum-l2) — a subset of TVS, just the stablecoin portion. - Profit / revenue (/answers/most-profitable-ethereum-l2) — what the chain earns from its activity, not what it holds. TVS leadership is sticky because liquidity is sticky — DeFi protocols, market makers, and large depositors don't move easily. The chains that built early TVS in 2022–2024 (Arbitrum, Base, OP Mainnet) tend to retain it. Methodology and data sources How the answer is derived: 1. Pull the master chain catalogue and filter to chains where , , and the chain key is not on the explicit non-L2 list below. 2. For each L2 in the universe, pull . The endpoint exposes a daily timeseries with columns — we resolve the USD column by name and take the latest day's value as the current TVS. 3. The 30-day change column uses the same series sampled 30 days back (point-in-time, since TVS is a stock metric). 4. Sort chains descending by current TVS and take the top 10. 5. Sum across the L2 universe to produce the ecosystem total used for share calculations. All values shown above were generated on 2026-05-27 UTC. Data licensed CC BY-NC 4.0. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Several supporters operate L2 chains that appear in the rankings above. Ranks are computed mechanically from public API data — chains don't pay for inclusion or placement. Full list of supporters: growthepie.com/donate. Cross-check this answer. Independent TVS / TVL data sources for Ethereum L2s include DefiLlama's chains page and L2BEAT. Methodologies differ slightly between providers — what counts as "secured" (native ETH? bridged ETH only? LSTs? RWAs?) and the price feed used can produce 5–10% differences in absolute USD value. Rank order and direction of movement should agree. Which chains are included? The list of 25 chains is computed automatically from and refreshed when growthepie adds or removes coverage: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Ethereum mainnet — it's Layer 1, not Layer 2. - Polygon PoS — a sidechain with its own validator set, not an L2. - Aggregate keys ( , ) — not individual chains. ### FAQ ### Which Ethereum L2 has the most total value secured (TVS)? As of 2026-05-27 UTC, the largest Ethereum L2 by total value secured is **Arbitrum One ($15.87B, 46% of L2 TVS; +1.9% 30d)**. Full top 10 by TVS: 1. Arbitrum One ($15.87B, 46% of L2 TVS; +1.9% 30d); 2. Base Chain ($12.15B, 35% of L2 TVS; +1.7% 30d); 3. OP Mainnet ($1.53B, 4.4% of L2 TVS; -2.8% 30d); 4. Mantle ($1.44B, 4.2% of L2 TVS; -11.1% 30d); 5. Starknet ($625.74M, 1.8% of L2 TVS; +17.8% 30d); 6. Linea ($447.59M, 1.3% of L2 TVS; -9.1% 30d); 7. Ink ($413.63M, 1.2% of L2 TVS; -3.2% 30d); 8. World Chain ($330.09M, 1.0% of L2 TVS; +19.5% 30d); 9. ZKsync Era ($292.37M, 0.8% of L2 TVS; -11.1% 30d); 10. Celo ($271.19M, 0.8% of L2 TVS; +2.0% 30d). TVS is the dollar value of all assets bridged to or natively held on the chain. Live leaderboards: [growthepie.com/chains](https://www.growthepie.com/chains). ### What is TVS (total value secured)? **Total value secured** is the total USD value of all assets bridged into or natively held on an Ethereum L2 — stablecoins, ETH, ERC-20 tokens, wrapped assets, etc. Often used interchangeably with **TVL (total value locked)**, though TVS more accurately reflects what L2s do (they secure value on behalf of users, who can withdraw it back to Ethereum). growthepie's [chains directory](https://www.growthepie.com/chains) labels this metric "Total Value Secured" by default. ### How is TVS different from TVL? They're mostly the same metric, just labelled differently. **TVL ("locked")** is the older term, often used by DeFi protocols to describe assets deposited into smart contracts. **TVS ("secured")** is the newer label preferred for L2s — it emphasises that the chain *secures* the assets on behalf of users rather than locking them in a single application. The number is the same for an L2: total dollar value of assets the chain holds. growthepie's API uses the `tvl` endpoint name for both. ### What's the total TVS across all Ethereum L2s? As of 2026-05-27 UTC, the combined TVS across the 25-chain Ethereum L2 universe is approximately **$34.56B**. The leader **Arbitrum One ($15.87B, 46% of L2 TVS; +1.9% 30d)** alone accounts for a significant share. See the full top-10 table for share breakdown. ### Which Ethereum L2 had the biggest TVS increase in the last 30 days? See the "30d change" column in the top-10 table above — chains with positive change pulled in net new capital, chains with negative change saw outflows. For a dedicated growth ranking that's independent of absolute size, see [/answers/fastest-growing-ethereum-l2](/answers/fastest-growing-ethereum-l2). ### How is TVS calculated? growthepie aggregates per-chain TVS from on-chain bridge contracts and asset balances. The methodology follows the same patterns used by [DefiLlama](https://defillama.com/) and [L2BEAT](https://l2beat.com) — assets are priced in USD at current market rates and summed across all bridged tokens. Native ETH balances on the L2 are included. Each chain's number is the latest daily snapshot. ### Why is one chain's TVS dominant? TVS concentrates around chains that have **(a)** been live for longest, **(b)** built deep DeFi liquidity, and **(c)** captured institutional capital. The dominant L2 today (typically Arbitrum or Base, depending on cycle) accumulated TVS over multiple years; newer chains start with low TVS and grow it via incentive programs, native app deployments, and bridge integrations. TVS leadership is sticky — once liquidity concentrates somewhere it doesn't move easily. ### Is Polygon PoS counted as an L2 here? No. Polygon PoS is a sidechain with its own validator set and is excluded from these L2 rankings — matching the rest of the L2 answer pages on growthepie. Polygon zkEVM is a ZK rollup and is included. See [/answers/l2-vs-sidechain](/answers/l2-vs-sidechain) for the full L2-vs-sidechain distinction. ### How many L2s are included? 25 chains. The full list (computed on 2026-05-27 UTC from growthepie's master chain catalogue) is: arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. ### Where does this data come from? Per-chain TVS values come from growthepie's per-chain TVL endpoint (`/v1/metrics/chains/{chain}/tvl.json`) — the same endpoint that backs the [chains directory](https://www.growthepie.com/chains) and the "Total Value Secured" column there. We take the latest daily USD value as the current TVS and the value 30 days prior as the comparison baseline. L2 membership comes from `master.json` (chains where `bucket !== "Layer 1"` and `chain_type` indicates an Ethereum rollup or validium). Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. No editorial overrides. ### How is "Ethereum L2" defined here? An Ethereum Layer 2 is a chain that derives security from Ethereum by posting transaction data and/or state to Ethereum mainnet. This includes optimistic rollups, ZK rollups, and Validiums. Sidechains (independent validator sets, like Polygon PoS) are excluded. See [/answers/l2-vs-sidechain](/answers/l2-vs-sidechain) for the full definition. ### Where can I see live TVS data? growthepie's [chains directory](https://www.growthepie.com/chains) shows every tracked chain with its current TVS sortable. Per-chain pages (e.g. [/chains/arbitrum](https://www.growthepie.com/chains/arbitrum)) include TVS timeseries charts. For independent verification see [DefiLlama's L2 page](https://defillama.com/chains) and [L2BEAT](https://l2beat.com). ### Why might TVS differ between providers? Different providers use different asset price feeds, different methodology for what counts as "secured" (does native ETH count? bridged ETH only? LSTs?), and different chain coverage. growthepie, DefiLlama, and L2BEAT broadly agree on rank order but the exact dollar values can differ by 5–10%. When in doubt, compare on **direction and rank** rather than exact figures. ### How does TVS compare across other metrics? TVS measures **capital weight** — how much value is parked on the chain. It's different from activity (transaction count, throughput, active addresses), which measures *flow* not *stock*. A chain can be high-TVS but low-activity (capital parked but not moving) or vice versa. For activity rankings see [/answers/most-used-ethereum-l2](/answers/most-used-ethereum-l2). For stablecoin-specific value, see [/answers/most-stablecoin-activity-ethereum-l2](/answers/most-stablecoin-activity-ethereum-l2). --- ## Which Ethereum L2 is the most profitable? URL: https://www.growthepie.com/answers/most-profitable-ethereum-l2 ### Direct answer By profit over the last 30 days, the most profitable Ethereum L2 is **Base Chain** at $3.58M ($9.69M over 90 days; $180.95M all-time). The L2 ecosystem collectively earned $4.64M in the last 30 days. Top 10: 1. Base Chain (30d: $3.58M; 90d: $9.69M; all-time: $180.95M); 2. Arbitrum One (30d: $560.1k; 90d: $1.80M; all-time: $79.44M); 3. Starknet (30d: $161.2k; 90d: $437.8k; all-time: $36.39M); 4. Celo (30d: $145.1k; 90d: $198.0k; all-time: $721.8k); 5. MegaETH (30d: $69.3k; 90d: $168.3k; all-time: $3.75M); 6. OP Mainnet (30d: $35.4k; 90d: $123.7k; all-time: $30.32M); 7. Linea (30d: $32.8k; 90d: $116.9k; all-time: $36.42M); 8. ZKsync Era (30d: $17.3k; 90d: $49.7k; all-time: $27.82M); 9. World Chain (30d: $14.7k; 90d: $22.6k; all-time: $718.8k); 10. Ink (30d: $11.5k; 90d: $33.5k; all-time: $265.7k). Data: 2026-05-27 UTC. Profit = revenue (fees) − L1 settlement cost (rent). Live leaderboards: growthepie.com/economics. ### Full answer Short answer (data 2026-05-27 UTC): Base Chain ($3.58M) is the most profitable Ethereum L2 over the last 30 days. The full top 10: 1. Base Chain (30d: $3.58M; 90d: $9.69M; all-time: $180.95M); 2. Arbitrum One (30d: $560.1k; 90d: $1.80M; all-time: $79.44M); 3. Starknet (30d: $161.2k; 90d: $437.8k; all-time: $36.39M); 4. Celo (30d: $145.1k; 90d: $198.0k; all-time: $721.8k); 5. MegaETH (30d: $69.3k; 90d: $168.3k; all-time: $3.75M); 6. OP Mainnet (30d: $35.4k; 90d: $123.7k; all-time: $30.32M); 7. Linea (30d: $32.8k; 90d: $116.9k; all-time: $36.42M); 8. ZKsync Era (30d: $17.3k; 90d: $49.7k; all-time: $27.82M); 9. World Chain (30d: $14.7k; 90d: $22.6k; all-time: $718.8k); 10. Ink (30d: $11.5k; 90d: $33.5k; all-time: $265.7k). The L2 ecosystem collectively earned $4.64M in profit over the last 30 days; $418.54M all-time across 25 tracked chains. Profit = chain revenue (fees collected from users) − L1 settlement cost (rent paid to Ethereum mainnet to post batches and proofs). Recomputed daily from growthepie's per-chain profit endpoint. Top 10 Ethereum L2s by profit As of 2026-05-27 UTC, Base Chain is the most profitable Ethereum L2 with $3.58M in profit over the last 30 days ($9.69M over 90 days; $180.95M all-time). Ecosystem total last 30 days: $4.64M; all-time: $418.54M. What "profit" means for an L2 An Ethereum L2 has two sides to its economics: - Revenue. Every L2 transaction pays a fee in the L2's gas token (usually ETH). The chain operator (the sequencer) collects that fee. - Cost. Periodically the L2 posts its transaction data and proofs to Ethereum mainnet. That costs L1 gas — paid in ETH to Ethereum validators. After Dencun (March 2024) this cost dropped sharply because L2s now use blobs instead of calldata. After Fusaka (December 2025) it dropped again. Profit = Revenue − Cost. A positive number means the chain is earning more from users than it pays Ethereum mainnet. A negative number means the chain is operating at a loss — usually because of an incentive program subsidising user fees, but sometimes structural. Why does this ranking matter? Profitability is the cleanest single measure of whether an L2 is operating as a sustainable business. Chains that consistently lose money are either burning down a treasury, running on grants, or planning to recoup later by capturing market share. Chains that consistently earn money are self-sustaining and don't depend on external funding to keep running. For long-term users and developers, this is a useful health check on which chains will still be around in three years. Caveats: - Profit isn't the only goal. Some chains intentionally run at a loss to subsidise user growth (and treasury growth from token appreciation, when applicable). - Profit isn't evenly captured. Where the profit goes varies — some chains return it to ecosystem incentives, others retain it as a foundation or DAO treasury, others split it with token holders. growthepie tracks the gross number, not the downstream allocation. Methodology and data sources How the answer is derived: 1. Pull the master chain catalogue and filter to L2s. 2. For each L2, pull . The endpoint exposes a daily timeseries of rows. 3. Sum the USD column over the last 30 days, last 90 days, and the entire series (all-time). 4. Rank chains by 30-day profit descending. Negative-profit chains still appear, just at the bottom. All values shown were generated on 2026-05-27 UTC. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Profit values are computed mechanically from public API data — chains don't influence the ranking. Full list of supporters: growthepie.com/donate. Cross-check this answer. Independent L2 economics sources include L2BEAT's costs view and growthepie's own economics dashboard which breaks down revenue vs cost vs profit per chain with charts. Which chains are included? The list of 25 chains: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. Excluded: Ethereum mainnet (L1), Polygon PoS (sidechain), aggregate keys. ### FAQ ### Which Ethereum L2 is the most profitable? Over the last 30 days (data 2026-05-27 UTC), the most profitable Ethereum L2 is **Base Chain ($3.58M)**. Profit = revenue from fees collected minus L1 settlement cost (rent paid to Ethereum mainnet). Full top 10: 1. Base Chain (30d: $3.58M; 90d: $9.69M; all-time: $180.95M); 2. Arbitrum One (30d: $560.1k; 90d: $1.80M; all-time: $79.44M); 3. Starknet (30d: $161.2k; 90d: $437.8k; all-time: $36.39M); 4. Celo (30d: $145.1k; 90d: $198.0k; all-time: $721.8k); 5. MegaETH (30d: $69.3k; 90d: $168.3k; all-time: $3.75M); 6. OP Mainnet (30d: $35.4k; 90d: $123.7k; all-time: $30.32M); 7. Linea (30d: $32.8k; 90d: $116.9k; all-time: $36.42M); 8. ZKsync Era (30d: $17.3k; 90d: $49.7k; all-time: $27.82M); 9. World Chain (30d: $14.7k; 90d: $22.6k; all-time: $718.8k); 10. Ink (30d: $11.5k; 90d: $33.5k; all-time: $265.7k). The L2 ecosystem collectively earned **$4.64M** in the last 30 days, **$418.54M** all-time. Live leaderboards: [growthepie.com/economics](https://www.growthepie.com/economics). ### What is "L2 profit"? **Profit = fees collected from users − cost of posting data and proofs to Ethereum mainnet.** When a user transacts on an L2, they pay a fee in the L2's gas token (usually ETH). The chain operator (the sequencer) collects that fee. Periodically, the L2 has to post its transaction data (or proofs) back to Ethereum mainnet — and pay Ethereum's L1 gas fee to do so. The difference between what the chain collected and what it paid Ethereum is the chain's profit. A profitable L2 charges its users more than it costs to settle on Ethereum. ### Can L2 profit be negative? Yes, and several L2s have run at a loss at various points. A chain runs negative-profit when its L1 settlement cost exceeds the fees it collected — typically because (a) the chain runs an incentive program subsidising user fees, (b) the chain posts data more aggressively than its activity warrants, or (c) network demand drops below break-even. **Sustained negative profit is usually deliberate** — the chain is buying market share — but it can't go on forever. ### Which L2 has earned the most all-time? See the "All-time profit" column in the top-10 table above. The all-time leader is typically the L2 with the longest production history *and* the highest sustained activity — Arbitrum, OP Mainnet, and Base are perennial leaders. New L2s with high recent profit show up high on the 30-day ranking but lower on all-time. ### Which L2 has earned the most in the last 90 days? See the "90d profit" column in the top-10 table above. The 90-day window is a useful middle ground between recent (30d) and structural (all-time) views — it smooths over single-month volatility while still reflecting the current state of the chain. ### How does growthepie calculate L2 profit? Per-day, per-chain: **revenue** (total transaction fees collected by the L2, USD-denominated) minus **L1 settlement cost** (rent paid to Ethereum mainnet for posting batches and proofs). The per-chain profit endpoint (`/v1/metrics/chains/{chain}/profit.json`) exposes a daily timeseries of that net figure in both USD and ETH. We sum daily values over 30d / 90d / all-time windows to produce this ranking. ### How is L2 revenue different from L2 profit? **Revenue** is what the chain collects from users in fees. **Profit** is revenue minus what the chain pays Ethereum mainnet to settle. A chain can have high revenue but low profit if it pays a lot to Ethereum — which is exactly what happens to chains with high settlement frequency or pre-Dencun-style calldata posting. After Dencun (March 2024), L1 settlement costs dropped sharply via blobs, and L2 profit margins widened. Fusaka (December 2025) widened them again. ### Where does L2 profit go? To the chain operator — typically a foundation, company, or DAO. Some chains rebate part of their profit to ecosystem incentives (Optimism's RetroPGF, Base's ecosystem grants); some operate primarily as a business and retain the profit. The mechanics vary chain-to-chain; growthepie doesn't track downstream usage of L2 profit. ### Why might an L2 with lots of activity not be very profitable? Three common reasons. **(1) Fee subsidy** — the chain runs an active incentive program rebating user fees, so collected revenue is depressed. **(2) Cheap-transaction profile** — the chain attracted high transaction count but with very low per-transaction fees (e.g. memecoin trading). **(3) Inefficient settlement** — the chain posts data to L1 more frequently than its activity warrants, paying L1 gas without proportional revenue. The top 10 profit ranking and the top 10 transactions ranking ([/answers/most-used-ethereum-l2](/answers/most-used-ethereum-l2)) are usually correlated but not identical. ### Is Polygon PoS counted as an L2 here? No. Polygon PoS is a sidechain with its own validator set and is excluded. See [/answers/l2-vs-sidechain](/answers/l2-vs-sidechain). ### How many L2s are included? 25 chains. The full list (computed on 2026-05-27 UTC): arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. A chain only appears in the ranking if growthepie has at least one day of profit data for it. ### Where does this data come from? Per-chain profit values come from growthepie's per-chain profit endpoint (`/v1/metrics/chains/{chain}/profit.json`). The endpoint exposes a daily timeseries of `(unix, USD, ETH)` rows; we sum the USD column over each window. L2 membership comes from `master.json`. Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. ### Where can I see this live? growthepie's [economics page](https://www.growthepie.com/economics) shows live revenue, costs, and profit per L2 with timeseries charts. Per-chain pages (e.g. [/chains/base](https://www.growthepie.com/chains/base)) include profit history in the economics section. ### How does this compare to "how much Ethereum mainnet earns from L2s"? Different sides of the same transaction. **L2 profit** = revenue − L1 cost. **Ethereum mainnet rent from L2s** = the L1 cost side, summed across L2s. The two are linked: every dollar of L1 settlement cost for an L2 is a dollar of revenue for Ethereum mainnet. See [/answers/ethereum-mainnet-revenue-from-l2s](/answers/ethereum-mainnet-revenue-from-l2s) for the mainnet side. --- ## How much does Ethereum mainnet earn from L2s? URL: https://www.growthepie.com/answers/ethereum-mainnet-revenue-from-l2s ### Direct answer Ethereum mainnet earns $1.2k per day, $46.3k per month, $131.6k over 90 days, and $333.36M all-time in settlement fees ("rent") from Ethereum L2s. Top contributors over the last 30 days: 1. Base Chain ($12.7k over 30d, 27% of L2 total); 2. World Chain ($5.6k over 30d, 12% of L2 total); 3. Taiko Alethia ($4.3k over 30d, 9.3% of L2 total); 4. Arbitrum One ($3.9k over 30d, 8.4% of L2 total); 5. Unichain ($3.4k over 30d, 7.4% of L2 total). Data: 2026-05-27 UTC. Live leaderboards: growthepie.com/economics. ### Full answer Short answer (data 2026-05-27 UTC): Ethereum L2s pay Ethereum mainnet approximately $1.2k per day, $46.3k per month, $131.6k over 90 days, and $333.36M all-time in L1 settlement fees. The top L2 contributor over the last 30 days is Base Chain ($12.7k over 30d, 27% of L2 total). Rent = the L1 gas fee an L2 pays when it posts batches (or proofs) to Ethereum mainnet. Summed across all tracked L2s, this is what Ethereum mainnet collectively earns from the rollup ecosystem. L2 → mainnet rent totals As of 2026-05-27 UTC, Ethereum L2s pay Ethereum mainnet $1.2k per day in settlement fees ("rent"), $46.3k per month, $131.6k over 90 days, and $333.36M all-time across 25 tracked L2s. - Daily (2026-05-27): $1.2k paid to Ethereum mainnet. - Last 30 days: $46.3k. - Last 90 days: $131.6k. - All-time (cumulative): $333.36M. Where does this money go? Every L1 transaction (including L2 batch posts) pays an Ethereum gas fee. Since EIP-1559 (August 2021), that gas fee splits into two parts: - Base fee — burned (permanently removed from circulation). This is where the bulk of L2 rent goes. It directly affects ETH supply and is one of the main reasons Ethereum has periods of deflation. See /answers/is-ethereum-deflationary. - Priority fee (tip) — paid to the validator that included the transaction in a block. So L2 rent doesn't go to "Ethereum" as an entity — it splits between ETH supply reduction (base fee burn) and validator income (priority fees). Both benefit ETH holders, but through different mechanisms. Which L2s contribute the most? Top contributors over the last 30 days (data 2026-05-27 UTC): 1. Base Chain ($12.7k over 30d, 27% of L2 total); 2. World Chain ($5.6k over 30d, 12% of L2 total); 3. Taiko Alethia ($4.3k over 30d, 9.3% of L2 total). Larger L2s by activity dominate — more L2 transactions mean more batches to settle mean more L1 fees paid. See the table for the full top 10 with each chain's share of the L2 ecosystem total. How upgrades have affected L2 rent Dencun (March 2024) — introduced blobs via EIP-4844. L2s switched from posting calldata (expensive) to blobs (10× cheaper). Per-L2-transaction settlement cost dropped sharply; L2 user fees dropped roughly proportionally; aggregate L1 rent revenue dropped but was partly offset by higher L2 activity. Pectra (May 2025) — doubled blob capacity (target 3→6 per block). More headroom for L2 activity without driving the blob fee market into escalation. Fusaka (December 2025) + BPO1/BPO2 — tripled blob target to 14 per block by January 2026. Also introduced EIP-7918, a floor on blob fees that ended the "zero-fee blob era" — ensuring L2 rent remains meaningful even during low-demand periods. See /answers/what-fusaka-upgrade-changed and /answers/what-pectra-upgrade-changed. Net effect of the 2024–2026 sequence: per-L2-transaction cost down sharply, aggregate L2 activity up sharply. Total rent to mainnet over time has been a noisy curve rather than monotonic growth or decline. Methodology and data sources How the answer is derived: 1. Pull the master chain catalogue and filter to L2s. 2. For each L2, pull — daily timeseries of L1 settlement fees in USD. 3. Sum each chain's daily values over the last 30 days, 90 days, and the entire series. Sum across L2s for ecosystem totals. 4. Rank contributors by 30-day rent paid (descending) for the leaderboard table. All values shown were generated on 2026-05-27 UTC. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Rent values are computed mechanically from public API data — chains don't influence the ranking. Full list of supporters: growthepie.com/donate. Cross-check this answer. ultrasound.money tracks total ETH burn (which includes L2 base-fee burn). L2BEAT's costs view tracks per-L2 data-posting costs. growthepie's economics dashboard shows the full revenue/cost/profit breakdown per chain. Which chains are included? The list of 25 chains: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. Excluded: Ethereum mainnet (L1), sidechains like Polygon PoS (don't settle to Ethereum), aggregate keys. ### FAQ ### How much does Ethereum mainnet earn from L2s? As of 2026-05-27 UTC, Ethereum L2s pay Ethereum mainnet approximately **$1.2k** per day in settlement fees, **$46.3k** per month, **$131.6k** over 90 days, and **$333.36M** all-time across 25 tracked L2s. The top contributor over the last 30 days is **Base Chain ($12.7k over 30d, 27% of L2 total)**. Live leaderboards: [growthepie.com/economics](https://www.growthepie.com/economics). ### What is "L2 rent" exactly? **Rent** is the L1 gas fee an L2 pays to Ethereum mainnet when it posts its batched transaction data (or proofs) back to L1. Every L2 has to settle on Ethereum periodically — that's what makes it an L2 — and each settlement transaction costs L1 gas, paid in ETH to Ethereum validators (post-Merge) and burned as base fee (post-EIP-1559). growthepie tracks this number per chain per day; the sum across all L2s is what Ethereum mainnet collectively earns from the rollup ecosystem. ### How much has Ethereum earned from L2s all-time? Approximately **$333.36M** cumulatively (data 2026-05-27 UTC, across 25 tracked L2s). This is the sum of every L1 settlement fee paid by every L2 since growthepie started tracking the metric. The number grew dramatically through 2023 (rollup adoption) but slowed after the **Dencun upgrade** in March 2024 — Dencun introduced blobs (EIP-4844), which made L2 settlement ~10× cheaper. Fusaka (December 2025) further reduced per-blob settlement cost. ### How much does Ethereum earn from L2s per day? On the latest day in growthepie's data (2026-05-27 UTC), Ethereum L2s collectively paid **$1.2k** to Ethereum mainnet in settlement fees. The number fluctuates day-to-day with L2 activity and L1 base fee. ### How much does Ethereum earn from L2s per month? Over the most recent 30 days ending 2026-05-27 UTC, Ethereum L2s collectively paid **$46.3k** to Ethereum mainnet. ### How much does Ethereum earn from L2s per quarter? Over the most recent 90 days ending 2026-05-27 UTC, Ethereum L2s collectively paid **$131.6k** to Ethereum mainnet. ### Which L2 pays Ethereum mainnet the most rent? Over the last 30 days (data 2026-05-27 UTC), the top three L2 contributors are 1. Base Chain ($12.7k over 30d, 27% of L2 total); 2. World Chain ($5.6k over 30d, 12% of L2 total); 3. Taiko Alethia ($4.3k over 30d, 9.3% of L2 total). Larger L2s by activity tend to dominate this list — more activity means more batches to post means more L1 fees paid. ### Does L2 rent go directly to Ethereum? It goes to Ethereum **validators** (the priority-fee portion) and to **burn** (the base-fee portion, since EIP-1559 in August 2021). Specifically: when an L2 posts a batch transaction to L1, it pays a base fee that's burned plus a priority tip that the proposer keeps. So L2 settlement fees are split between validator income and ETH supply contraction — which directly affects whether Ethereum is deflationary. See [/answers/is-ethereum-deflationary](/answers/is-ethereum-deflationary). ### How did Dencun affect L2 rent to mainnet? Dencun (March 2024) introduced **blobs** via EIP-4844 — a new transaction type optimised specifically for L2 data availability. Before Dencun, L2s posted their data as calldata, which is expensive. After Dencun, they post as blobs, which costs ~10× less. The effect: **per-L2-transaction L1 settlement cost dropped by an order of magnitude overnight**. L2 fees to users dropped roughly proportionally, and Ethereum mainnet's rent revenue from L2s dropped too — though the lower per-unit cost was partly offset by higher L2 activity. ### How did Fusaka affect L2 rent to mainnet? Fusaka (December 2025) tripled the blob target from 6 to 14 per block, and BPO2 raised it again in January 2026. More blob capacity = more L2 activity at the same per-blob price = more rent paid to mainnet in absolute terms, even at lower per-transaction cost. Fusaka also introduced **EIP-7918**, a floor on blob fees that prevented the "zero-fee blob era" — ensuring L2 rent stays meaningful even during quiet network periods. See [/answers/what-fusaka-upgrade-changed](/answers/what-fusaka-upgrade-changed). ### How is "rent paid" calculated? Per-chain, per-day: growthepie aggregates all L1 transactions made by each L2's posting contract (batch submitter, proof verifier, etc.) and sums the gas fees paid. The number is published as a daily USD value in `/v1/metrics/chains/{chain}/rent_paid.json`. This page sums per-chain daily values over windows (30d / 90d / all-time) and across the L2 universe to produce the totals. ### Is Polygon PoS counted as an L2 here? No. Polygon PoS is a sidechain — it doesn't pay rent to Ethereum mainnet because it doesn't settle to Ethereum. See [/answers/l2-vs-sidechain](/answers/l2-vs-sidechain) for the distinction. ### How many L2s are included? 25 chains. The full list (computed on 2026-05-27 UTC): arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. Sidechain exclusions: Polygon PoS. ### Where can I see this data live? growthepie's [economics dashboard](https://www.growthepie.com/economics) breaks down per-chain revenue, costs, and profit with timeseries charts. Per-chain pages (e.g. [/chains/base](https://www.growthepie.com/chains/base)) show rent paid history. [ultrasound.money](https://ultrasound.money/) tracks the burn side directly. ### How does this relate to L2 profitability? Two sides of the same transaction. **L2 rent to mainnet** = the L1 cost an L2 pays to settle. **L2 profit** = the L2's revenue minus that L1 cost. The two are linked: every dollar of rent paid is a dollar of cost for the L2 (and a dollar of revenue for Ethereum mainnet). See [/answers/most-profitable-ethereum-l2](/answers/most-profitable-ethereum-l2) for the L2-side ranking. --- ## Which Ethereum L2s are Stage 1 or Stage 2 rollups? URL: https://www.growthepie.com/answers/ethereum-l2-maturity-stages ### Direct answer growthepie's 25-chain Ethereum L2 universe is classified by maturity: 5 maturing, 1 developing, 9 emerging, 10 early phase. **Maturing** L2s (most decentralised on growthepie's scale): Arbitrum One, Base Chain, Ink, OP Mainnet, Starknet. **Developing** L2s: Mantle. growthepie's maturity scale is related to (but distinct from) L2BEAT's "Stage 0 / Stage 1 / Stage 2" framework and the Ethereum Foundation's per-L2 maturity assessment on ethereum.org — all three measure the same underlying decentralisation question with different emphases. Data: 2026-05-27 UTC. Live: growthepie.com/chains, l2beat.com, ethereum.org/en/layer-2. ### Full answer Short answer (data 2026-05-27 UTC): growthepie's 25-chain Ethereum L2 universe breaks down on growthepie's maturity scale as: Maturing (5): Arbitrum One, Base Chain, Ink, OP Mainnet, Starknet. Developing (1): Mantle. Emerging (9): Arbitrum Nova, Linea, Lisk, Loopring, Mode Network, Soneium, Taiko Alethia, World Chain, ZKsync Era. Early phase (10): Celo, Fraxtal, Gravity, Manta Pacific, MegaETH, Metis, Plume Network, Ronin, Scroll, Unichain. For the strict L2BEAT Stage 0/1/2 view see l2beat.com; for the Ethereum Foundation's own maturity assessment see ethereum.org/en/layer-2. Three related frameworks describe L2 decentralisation: L2BEAT's Stage 0/1/2 (strict, protocol-property-based), ethereum.org's maturity (the Ethereum Foundation's own per-L2 assessment, using the same word "maturity" growthepie uses), and growthepie's maturity (broader, incorporates chain age + ecosystem size + decentralisation in a single 0–3 ordinal). All three broadly agree on which chains are most mature; they aren't identical. The three frameworks Three widely-used frameworks describe "how decentralised is this Ethereum L2". They're parallel, not redundant — each emphasises a different angle of the same underlying question. 1. L2BEAT's Stage framework (strict, canonical for protocol decentralisation): - Stage 0 — training wheels on. Security council can override, fraud/validity proofs aren't live in production. Most newer L2s start here. - Stage 1 — proofs are live (protocol can verify state without a human), but a security council can still pause or override during emergencies. - Stage 2 — fully decentralised. No security-council override path. Very few L2s have reached this level as of 2026. 2. ethereum.org's maturity (the Ethereum Foundation's own per-L2 view): The ethereum.org Layer 2 page publishes the EF's maturity assessment for each tracked L2. It considers rollup Stage (drawing on L2BEAT), time live in production, total value secured, and ecosystem alignment with Ethereum. The EF uses the same word "maturity" that growthepie does — the frameworks are independent but the vocabulary overlaps. 3. growthepie's maturity scale (broader, used on growthepie): - Maturing (highest) — established L2s with multi-year production history, large ecosystem, and strong decentralisation properties. Typically L2BEAT Stage 1 or higher. - Developing — production-grade with meaningful activity, working toward full decentralisation. Often L2BEAT Stage 0 with a roadmap to Stage 1. - Emerging — newer or smaller production chains. Activity is growing. - Early phase — very young or experimental. Significant centralisation in the operator. How they relate: L2BEAT's Stage is the strictest — a binary check on specific protocol properties. ethereum.org's maturity is the EF's holistic view that incorporates L2BEAT's Stage plus production history and alignment. growthepie's maturity is broader still, with a single 0–3 ordinal that rolls in chain age and ecosystem size. The three correlate strongly but aren't identical — a chain can be growthepie "Maturing" while sitting at L2BEAT Stage 0 if it has long production history but hasn't shipped proofs yet. Every Ethereum L2 grouped by maturity Below is every tracked Ethereum L2 on growthepie, grouped by maturity level. Within each group chains are listed alphabetically. Maturing (5 chains) - Arbitrum One - Base Chain - Ink - OP Mainnet - Starknet Developing (1 chain) - Mantle Emerging (9 chains) - Arbitrum Nova - Linea - Lisk - Loopring - Mode Network - Soneium - Taiko Alethia - World Chain - ZKsync Era Early phase (10 chains) - Celo - Fraxtal - Gravity - Manta Pacific - MegaETH - Metis - Plume Network - Ronin - Scroll - Unichain Why does the stage matter? Practical consequence: what could go wrong. At lower maturity levels, the chain operator (or a small security council) can effectively pause, censor, or revert the chain. That's deliberate and reasonable for chains still proving out their bug-free operation — but it means the chain's trust model isn't just "Ethereum + the protocol", it's "Ethereum + the protocol + the team's good behaviour". At higher maturity levels, the chain operates increasingly close to "just the protocol" — Ethereum can verify state transitions cryptographically (or via fraud proofs) without trusting any single party. This is what users implicitly expect when they think of an L2 as "Ethereum but cheaper". For high-value use cases (DeFi treasuries, settlement, large NFT holdings), more mature = closer to Ethereum's actual trust model = lower extra risk. For low-value experiments and games, the practical difference may not matter. How chains progress The path from Stage 0 → Stage 1 → Stage 2 is risk-progressive, not lazy. Fraud-proof and validity-proof systems are among the most complex software anyone has ever shipped; bugs in them can mean total loss of funds. Most teams: 1. Launch with a centralised sequencer + security council (Stage 0). 2. Ship fraud or validity proofs to production while keeping a security-council backstop (Stage 1). 3. Retire the security council once the protocol has been proven over time (Stage 2). L2BEAT publishes per-chain roadmaps for these transitions; growthepie's field is reviewed periodically as chains progress through these milestones. Methodology and data sources Chain classifications on this page come from growthepie's master chain catalogue, specifically the field per chain. Values: , , , , . growthepie's classifications broadly align with L2BEAT's Stage assessment, ethereum.org's maturity view, and observable chain properties (age, ecosystem size, proof system status). For the canonical strict Stage 0/1/2 view, see l2beat.com. For the Ethereum Foundation's own maturity rankings, see ethereum.org/en/layer-2 — they publish per-L2 maturity using the same vocabulary growthepie uses. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Many supporters operate L2 chains at various maturity levels. Classifications on this page follow growthepie's published methodology and L2BEAT — no editorial bias. Full list of supporters: growthepie.com/donate. Which chains are included? All 25 chains in growthepie's L2 universe: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. Excluded: Ethereum mainnet (L1), Polygon PoS (sidechain), aggregate keys. ### FAQ ### Which Ethereum L2s are Stage 1 or Stage 2 rollups? Three related frameworks describe how decentralised an Ethereum L2 is: **L2BEAT's "Stages"** (Stage 0 = training wheels, Stage 1 = proofs live with security council backstop, Stage 2 = fully decentralised), **ethereum.org's "maturity"** (the Ethereum Foundation's own per-L2 maturity assessment), and **growthepie's "maturity"** scale (0–3 axis: Maturing / Developing / Emerging / Early phase). All three broadly correlate. As of 2026-05-27 UTC, growthepie's 25-chain Ethereum L2 universe breaks down as: **Maturing (5)**: Arbitrum One, Base Chain, Ink, OP Mainnet, Starknet. **Developing (1)**: Mantle. **Emerging (9)**: Arbitrum Nova, Linea, Lisk, Loopring, Mode Network, Soneium, Taiko Alethia, World Chain, ZKsync Era. **Early phase (10)**: Celo, Fraxtal, Gravity, Manta Pacific, MegaETH, Metis, Plume Network, Ronin, Scroll, Unichain. For the strict Stage 0/1/2 view see [L2BEAT](https://l2beat.com); for the EF's own assessment see [ethereum.org's Layer 2 page](https://ethereum.org/en/layer-2/). ### What is L2BEAT's "Stage" framework? A three-level classification (Stage 0 → Stage 1 → Stage 2) that measures **how decentralised a rollup actually is in production**. **Stage 0** = training wheels are on — the team has full control via a security council that can override the protocol, fraud or validity proofs aren't live in production. **Stage 1** = proofs are live (the protocol can validate or invalidate state transitions independently), but a security council can still pause or override during emergencies. **Stage 2** = fully decentralised — no security council overrides, the protocol stands alone. As of 2026 only a handful of L2s have reached Stage 1; very few have reached Stage 2. ### Does ethereum.org also classify L2 maturity? **Yes.** [ethereum.org's Layer 2 page](https://ethereum.org/en/layer-2/) publishes the Ethereum Foundation's own per-L2 **maturity** assessment, using the same word ("maturity") that growthepie uses. The EF framework considers factors like rollup stage, time live in production, total value secured, and ecosystem alignment with Ethereum. ethereum.org's "maturity", L2BEAT's "Stage", and growthepie's "maturity" are all parallel frameworks for the same underlying question — they're not identical, but they broadly agree on which L2s are most production-ready. ### How do growthepie's "maturity", ethereum.org's "maturity", and L2BEAT's "Stage" differ? All three measure how production-ready a rollup is, but they emphasise different things. **L2BEAT's Stage** is the strictest — it requires specific cryptographic and governance properties (fraud or validity proofs in production, no override mechanism for Stage 2, etc.) and reads as a binary protocol-property check. **ethereum.org's maturity** is the Ethereum Foundation's own assessment, incorporating L2BEAT's stage, time live, TVS, and Ethereum alignment. **growthepie's maturity** is broader still — it incorporates chain age, ecosystem size, production history, and decentralisation in a single 0–3 ordinal. The three frameworks roughly correlate (most L2s growthepie marks "Maturing" are L2BEAT Stage 1+ and rank highly on ethereum.org's maturity) but they're not identical. **For canonical protocol-decentralisation, use L2BEAT. For the EF's view, use ethereum.org. growthepie's scale is what we display on the site.** ### Which Ethereum L2s are at the "Maturing" stage? 5 chains: **Arbitrum One, Base Chain, Ink, OP Mainnet, Starknet**. "Maturing" is the highest level on growthepie's scale and typically maps to L2BEAT Stage 1 or beyond. These chains have been in production for years, have substantial value secured, and have shipped (or are shipping) the cryptographic/governance properties that allow them to operate with reduced training wheels. ### Which Ethereum L2s are at the "Developing" stage? 1 chains: **Mantle**. "Developing" L2s are typically in production with meaningful activity and TVS, but haven't yet shipped the full set of decentralisation properties needed to reach growthepie's "Maturing" tier. Many of them are on L2BEAT Stage 0 with a stated roadmap to Stage 1. ### Which Ethereum L2s are at the "Emerging" stage? 9 chains: **Arbitrum Nova, Linea, Lisk, Loopring, Mode Network, Soneium, Taiko Alethia, World Chain, ZKsync Era**. "Emerging" L2s are newer or smaller production chains — often launched within the last 1–2 years, often with growing but not yet large activity. They tend to be on L2BEAT Stage 0 with security-council oversight as the primary safety mechanism. ### Which Ethereum L2s are at the "Early phase"? 10 chains: **Celo, Fraxtal, Gravity, Manta Pacific, MegaETH, Metis, Plume Network, Ronin, Scroll, Unichain**. "Early phase" chains are very young or very experimental — the chain operator retains significant control via a centralised sequencer and governance, fraud/validity proofs may not yet be deployed in production. Users should treat these chains as higher-risk and follow the chain's own roadmap toward more decentralised operation. ### Does the maturity level actually matter for users? Yes, especially for high-value use cases. At lower maturity levels the chain operator (or a small security council) can effectively pause, censor, or revert the chain in an emergency. This is **deliberate and reasonable** for chains still ironing out bugs — but it means the chain's trust model isn't just "Ethereum + the protocol", it's "Ethereum + the protocol + the team's good behaviour". For deposits of meaningful value, the more mature the L2, the closer its trust model is to Ethereum's. For small experiments or NFT projects, the practical difference may not matter. ### Why do some L2s deliberately stay at Stage 0? Engineering caution. Fraud-proof and validity-proof systems are some of the most complex software anyone has ever shipped — bugs in them can mean total loss of funds. Most teams ship the chain first with a security-council backstop (so they can override in an emergency), prove out the system in production, then retire the security council once they're confident the protocol can stand alone. Going from Stage 0 → Stage 1 → Stage 2 is **risk-progressive**, not laziness. L2BEAT publishes per-chain roadmaps for the stage transition. ### Does ZK vs Optimistic affect the stage? Not directly. Both ZK and Optimistic rollups can be at any stage — the classification is independent of the proof system. **L2BEAT cares about whether proofs are live in production**, regardless of whether they're validity proofs (ZK) or fraud proofs (Optimistic). What matters is: can the protocol verify a state transition without a human in the loop? See [/answers/zk-vs-optimistic-rollup](/answers/zk-vs-optimistic-rollup) for the proof-system comparison. ### Where does this data come from? growthepie's maturity classification comes from each chain's `maturity` field in [master.json](https://api.growthepie.com/v1/master.json). Values are: `3_maturing`, `2_developing`, `1_emerging`, `0_early_phase`, or `NA` (not classified). We don't make editorial calls — the classifications follow growthepie's published methodology and broadly align with **L2BEAT's Stage** assessment and **ethereum.org's maturity** view. For the strict L2BEAT Stage view see [l2beat.com](https://l2beat.com); for the Ethereum Foundation's own maturity rankings see [ethereum.org/en/layer-2](https://ethereum.org/en/layer-2/). ### How many L2s are included? 25 chains. Full list: arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. Ethereum mainnet is excluded because it's Layer 1, not Layer 2. ### Where can I see this data live? growthepie's [chains directory](https://www.growthepie.com/chains) shows each chain's maturity classification alongside its activity metrics. [L2BEAT's stages page](https://l2beat.com) is the canonical reference for protocol-level Stage 0/1/2 classification. [ethereum.org's Layer 2 page](https://ethereum.org/en/layer-2/) publishes the Ethereum Foundation's own per-L2 maturity assessment. --- ## What is Data Availability (DA) and what are the differences between providers? URL: https://www.growthepie.com/answers/what-is-data-availability ### Direct answer Data Availability (DA) is the guarantee that a chain's transaction data is publicly accessible so anyone can reconstruct the chain's state — foundational to how Ethereum L2s work. The main DA providers for Ethereum L2s are Ethereum mainnet (via blobs since the Dencun upgrade in March 2024), Celestia, EigenDA, and Avail. They differ on cost, throughput, and security model. Live per-MB costs from growthepie: As of 2026-05-27 UTC (30-day average), per-megabyte data-availability costs are: Celestia: $0.0351 per MB (30d avg); EigenDA: $0.0464 per MB (30d avg); Ethereum DA (blobs): $0.169 per MB (30d avg). Note: growthepie's per-MB cost endpoint currently covers Celestia and EigenDA only; Ethereum blobs and Avail are tracked elsewhere (or not yet tracked). Live tracker: growthepie.com/data-availability. Data: 2026-05-27 UTC. ### Full answer Short answer: Data Availability (DA) is the guarantee that a chain's transaction data is publicly accessible, so anyone can reconstruct the chain's state. It's foundational to how Ethereum L2s work — without DA, an L2 can prove its state is valid but you can't verify that or withdraw funds. The main DA providers are Ethereum mainnet (blobs), Celestia, EigenDA, and Avail. They differ mainly on cost, throughput, and security model (Ethereum DA inherits Ethereum; alt-DA has separate trust assumptions). Live cost comparison: As of 2026-05-27 UTC (30-day average), per-megabyte data-availability costs are: Celestia: $0.0351 per MB (30d avg); EigenDA: $0.0464 per MB (30d avg); Ethereum DA (blobs): $0.169 per MB (30d avg). Live per-DA-layer metrics — blobs posted, data per day, fees paid, fees per MB — live at growthepie.com/data-availability. What is Data Availability? When an L2 batches up its transactions and settles to Ethereum, it has to give the world access to the underlying transaction data. Without that: - Users can't withdraw. If the L2 disappears, you need the transaction data to prove you own funds on Ethereum. - Optimistic rollups can't verify state. Fraud proofs need to replay the transactions; if the data is missing, fraud can't be challenged. - Anyone wanting to run a node can't catch up. A new validator or auditor needs the historical data to verify the chain. Data Availability is the guarantee that this data is publicly accessible for a long enough window that any honest party can grab it. Without DA, even a cryptographically valid L2 becomes effectively a custodial chain — you have to trust whoever's holding the data. The four main DA providers For Ethereum L2s, four DA options dominate as of 2026: 1. Ethereum DA (blobs). Since Dencun (March 2024, EIP-4844), Ethereum mainnet provides DA through blobs — large 128 KB data attachments priced separately from regular L1 gas. Blobs live on Ethereum's consensus layer for 18 days. Fusaka (December 2025) added PeerDAS which lets nodes verify blob availability via sampling, enabling further capacity increases. Maximum security (inherits Ethereum), most expensive per byte. 2. Celestia. Standalone proof-of-stake blockchain dedicated to DA. Mainnet October 2023. L2s post data to Celestia (paying TIA tokens); Celestia validators sample blocks to confirm availability. Cheap, high-throughput, separate trust assumption (Celestia validators, not Ethereum). 3. EigenDA. Built on EigenLayer — Ethereum's restaking ecosystem. Restaked ETH backs the DA layer's economic security. Cheap, high-throughput, ETH-backed (but not pure Ethereum consensus). Used by chains like MegaETH. 4. Avail. Another standalone modular DA layer, similar in architecture to Celestia. Uses KZG commitments and data availability sampling. Cheap, high-throughput, separate trust assumption. Provider comparison Practical comparison of the trade-offs that matter most: | Provider | Security model | Cost ($/MB, 30d avg) | Throughput | Native token | Mainnet since | |---|---|---|---|---|---| | Ethereum DA (blobs) | Inherits Ethereum (highest) | $0.169 | Capped by blob target | ETH | March 2024 | | EigenDA | Restaked ETH (slashable) | $0.0464 | High | ETH (restaked) | 2024 | | Celestia | Own PoS validators | $0.0351 | High | TIA | October 2023 | | Avail | Own PoS validators | unavailable | High | AVAIL | 2024 | Costs are live 30-day averages from growthepie's endpoint (data 2026-05-27 UTC). Where a value reads "unavailable", growthepie's per-MB cost endpoint doesn't expose that provider today — see the methodology FAQ. The pattern: Ethereum DA is the conservative, max-security option but the most expensive per byte. The alt-DA providers are typically much cheaper but introduce a separate trust assumption (the alt-DA layer's validators or restakers). L2BEAT classifies chains posting to alt-DA as Validiums (ZK + alt-DA) or Optimiums (Optimistic + alt-DA) rather than pure rollups, reflecting the extra trust assumption. How Data Availability Sampling works DAS is the technical breakthrough that makes high-throughput DA practical. Instead of every node downloading every byte of every block, nodes each download a small random sample. The data is erasure-coded so any subset of pieces above a threshold can reconstruct the whole. If enough nodes sample randomly, the network collectively confirms any missing piece would have been detected. Celestia shipped DAS first as a production design. Ethereum added it via PeerDAS (EIP-7594) in Fusaka (December 2025) — see /answers/what-fusaka-upgrade-changed. PeerDAS is what makes the Fusaka and post-Fusaka blob target increases (target 6 → 14 per block by January 2026) safe at the network level. Which Ethereum L2s use which DA? Most major L2s use Ethereum DA (blobs) — Arbitrum, OP Mainnet, Base, zkSync Era, Linea, Scroll, etc. Several chains use alt-DA for cost: MegaETH on EigenDA; Manta Network has used Celestia; many Polygon CDK chains can choose between options at deployment. Many chains are also hybrid — they post critical state to Ethereum and bulk data to alt-DA. growthepie's data availability tracker shows the live per-chain DA mapping. How to pick a DA provider (if you're building an L2) The answer is rarely "the cheapest" or "the most secure" — it's "the right trade-off for the use case": - DeFi, settlement, treasury-grade L2s — Ethereum DA. The premium is worth the strongest possible security guarantee, and DeFi users are sensitive to trust assumptions. - Gaming, social, NFT, high-throughput consumer L2s — alt-DA. The cost difference is meaningful at scale, and the typical use case doesn't need Ethereum's full security profile. - Hybrid chains — split the difference, posting some data to Ethereum and bulk to alt-DA. There's no "right" choice — it's an engineering trade-off each chain team makes for their specific user base. Methodology and data sources Conceptual material on this page is drawn from the published EIPs (EIP-4844 for blobs, EIP-7594 for PeerDAS), the Celestia / EigenDA / Avail whitepapers, and L2BEAT's DA risk framework. Live per-DA-layer metrics come from growthepie's DA endpoints ( , , and the per-DA endpoints). Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Several supporters are DA providers or chains using specific DA providers. This page presents technical facts — provider descriptions follow the providers' own published material and L2BEAT's risk classifications. Full list of supporters: growthepie.com/donate. Cross-check this answer. L2BEAT has detailed DA risk analysis per L2. ethereum.org's roadmap covers Ethereum's own DA approach. The DA providers' websites — celestia.org, eigenda.xyz, availproject.org — have their own whitepapers and live dashboards. ### FAQ ### What is Data Availability (DA) and what are the differences between providers? **Data Availability (DA)** is the guarantee that a chain's transaction data is **publicly available** — anyone who wants to reconstruct the chain's state can do so. It's the foundation L2 rollups stand on: an L2 can prove its state is *valid*, but unless the data is available, users can't verify that or withdraw their funds. The main DA providers for Ethereum L2s are **Ethereum mainnet (via blobs since Dencun)**, **Celestia**, **EigenDA**, and **Avail**. They differ on cost, throughput, security model, and who finalises the data. growthepie tracks them all at [growthepie.com/data-availability](https://www.growthepie.com/data-availability). ### Why does data availability matter for an L2? An L2's entire security model depends on data being available. Without it, three things break: **(1) Withdrawal safety** — users can't prove they own funds on the L2 if the data has disappeared. **(2) Fraud-proof construction** — Optimistic rollups need the underlying transaction data to construct fraud proofs. **(3) State verification** — any independent party that wants to verify the L2's state needs the data to do so. If a DA layer fails to make data available, even a "valid" L2 becomes effectively a custodial chain — you have to trust whoever holds the data. ### What is "Ethereum DA" (blobs)? Since the **Dencun upgrade (March 2024, EIP-4844)**, Ethereum mainnet itself provides DA through a new data type called **blobs** — large data attachments (~128 KB each) priced separately from regular L1 gas. Blobs live on Ethereum's consensus layer for ~18 days, long enough for any honest party to download and store them. After 18 days the chain still trusts that someone, somewhere, has the data; this is the same "weak subjectivity" assumption Ethereum already uses for consensus history. **Fusaka (December 2025)** added **PeerDAS** which lets nodes verify blob availability without downloading entire blobs, enabling further capacity increases. **Verdict: maximum security (inherits Ethereum), most expensive per byte, capacity capped by blob target.** ### What is Celestia? **Celestia** is a dedicated **modular DA layer** — a standalone proof-of-stake blockchain whose only job is publishing and validating data blobs for other chains. Celestia launched mainnet in October 2023. L2s can post their data to Celestia instead of (or alongside) Ethereum, paying CelestiaTIA tokens for the data. Celestia validators sample a fraction of each block to confirm data availability — the same principle as Ethereum's PeerDAS, but Celestia shipped it first as a production design. **Verdict: cheap per byte, high throughput, separate trust assumption (you trust Celestia validators, not Ethereum validators).** ### What is EigenDA? **EigenDA** is a DA service built on **EigenLayer** — Ethereum's restaking ecosystem. Users restake their ETH (or LSTs) to opt into providing DA services; the restaked ETH backs the cryptoeconomic security of the DA layer. EigenDA inherits a large portion of Ethereum's economic security without paying Ethereum gas for blob storage. Launched mainnet in 2024. **Verdict: cheap per byte, high throughput, ETH-backed security (you trust the restakers/operators with slashable ETH at risk, not pure Ethereum consensus).** Used by chains like MegaETH and others. ### What is Avail? **Avail** (originally part of Polygon, now an independent project) is another **modular DA layer**, similar in architecture to Celestia. It uses KZG commitments and data availability sampling to allow light clients to verify data is available without downloading it. Mainnet launched in 2024. **Verdict: cheap per byte, high throughput, separate trust assumption (you trust Avail validators).** ### Which DA provider is cheapest? **Live data (30-day average, data 2026-05-27 UTC):** Celestia: **$0.0351** per MB. EigenDA: **$0.0464** per MB. Ethereum DA (blobs): **$0.169** per MB. Avail: **unavailable** per MB. Where a value reads "unavailable", growthepie's per-megabyte cost endpoint doesn't yet expose that provider — see the methodology FAQ for which providers are tracked where. **Pricing fluctuates** with each provider's native fee market (Celestia prices in TIA, EigenDA in ETH at restaker rates, Avail in AVAIL, Ethereum DA in ETH via the blob base fee), so the 30-day average smooths over short-term swings. Live charts: [growthepie.com/data-availability](https://www.growthepie.com/data-availability). ### Which DA provider is most secure? **Ethereum DA (blobs) is the highest security** — same validators, same consensus, same economic backing as Ethereum mainnet itself. **EigenDA** comes second because it's backed by restaked ETH (slashing risk = real cost to attack). **Celestia and Avail** have their own dedicated validators and economic security, which is robust but lower than Ethereum's in absolute terms. For chains that need the strongest possible security guarantee (DeFi treasuries, settlement-critical assets), Ethereum DA is the conservative choice. For chains optimising for cheap, high-throughput data (gaming, social), alt-DA is the rational trade-off. ### Are chains using alt-DA still "Ethereum L2s"? It depends on definition. **L2BEAT classifies chains that post data to non-Ethereum DA layers as "Validiums" or "Optimiums"** rather than as pure rollups — the security model has an extra trust assumption (the alt-DA layer). growthepie and most of the ecosystem still consider them "Ethereum L2s" in the colloquial sense because they settle state to Ethereum, even if their data lives elsewhere. The technical term to be precise is **Validium** (ZK + alt-DA) or **Optimium** (Optimistic + alt-DA). See [/answers/zk-vs-optimistic-rollup](/answers/zk-vs-optimistic-rollup). ### How does Data Availability Sampling (DAS) work? **Data Availability Sampling** lets a node confirm that a large block is fully available *without* downloading the whole thing. Each node downloads a small random sample of pieces; if enough nodes each sample randomly, collectively they can verify any missing piece would have been detected. The underlying mathematical trick is **erasure coding** — the data is expanded so that any subset of pieces above a threshold can reconstruct the whole. DAS is the breakthrough that makes high-throughput DA practical: instead of every node needing a copy of every byte, the network collectively verifies availability at much lower per-node cost. Celestia shipped DAS first; Ethereum added it via **PeerDAS (EIP-7594) in the Fusaka upgrade** (December 2025). ### How long does data have to be available? Long enough that any honest party who wants the data has time to download and store it. Ethereum keeps blobs for **~18 days** on the consensus layer — after that the chain still trusts that someone, somewhere, has the data archived. Alt-DA layers have similar windows (Celestia keeps data for ~30 days, EigenDA configurable). After the on-chain window expires, the security model becomes "weak subjectivity" — you trust archives (the chain team, light-client archivists, third-party storage) to retain the data. This is the same assumption Ethereum itself uses for old block data older than the chain's reorg boundary. ### Which Ethereum L2s use which DA provider? Most major L2s use **Ethereum DA (blobs)** — Arbitrum, OP Mainnet, Base, zkSync Era, Linea, Scroll, etc. Some chains use **alt-DA** for cost reasons — MegaETH uses EigenDA, Manta Network has used Celestia, several Polygon CDK chains can choose between options. Many chains are also **hybrid** — they post critical state to Ethereum and bulk data to alt-DA. growthepie's [data availability tracker](https://www.growthepie.com/data-availability) shows which chain uses which DA, plus per-DA-layer activity charts. ### How does growthepie classify DA layers? In growthepie's [master.json](https://api.growthepie.com/v1/master.json), each chain has a `da_layer` field identifying which DA provider it uses. There's also a top-level `da_layers` object describing each provider. growthepie publishes per-DA-layer dashboards under [/data-availability](https://www.growthepie.com/data-availability) with metrics like blobs per block, data posted (MB/day), fees paid (USD), and fees per megabyte (cost efficiency). ### Why does the cost comparison show "unavailable" for some providers? growthepie's `/v1/da_metrics/fees_per_mbyte.json` endpoint currently exposes per-megabyte fees for **Celestia and EigenDA**, but not yet for **Ethereum DA (blobs)** or **Avail**. Ethereum blob fees and total Avail fees are tracked elsewhere on growthepie (via `da_overview.json` for Ethereum blobs; Avail isn't in growthepie's DA coverage as of 2026), but a clean apples-to-apples per-MB number isn't served by the same endpoint. We don't fabricate values — where a number isn't directly available, the page reads "unavailable". As the upstream endpoint expands coverage, this page will automatically pick up the additional values. ### Where can I see live DA data? growthepie's [data availability section](https://www.growthepie.com/data-availability) has dashboards for each tracked DA layer — Ethereum, Celestia, EigenDA, Avail — with charts for blob count, data posted, fees paid, and per-megabyte cost. Per-chain pages also show which DA each L2 uses. [L2BEAT](https://l2beat.com) has architectural detail on each chain's DA choice and the resulting security profile. --- ## What was The Merge and how did it change Ethereum? URL: https://www.growthepie.com/answers/what-was-the-merge ### Direct answer The Merge activated on September 15, 2022 and transitioned Ethereum from Proof-of-Work (miners) to Proof-of-Stake (validators staking ETH). Three immediate effects: new ETH issuance dropped ~87% (from ~13,000 ETH/day of mining rewards to ~1,700 ETH/day of staking rewards), energy consumption fell ~99.95% (no more mining hardware), and the conditions for ETH deflation were created — combined with EIP-1559's base-fee burn already in place, net-negative ETH issuance became possible during periods of high network activity. The Merge did NOT change transaction fees, throughput, or smart contracts — it was a pure consensus-layer change. It's widely regarded as one of the largest live-system upgrades in software engineering history. ### Full answer Short answer: The Merge activated on September 15, 2022, transitioning Ethereum from Proof-of-Work to Proof-of-Stake. Issuance dropped 87%, energy use dropped 99.95%, and combined with EIP-1559's burn (already active), the conditions for ETH net-deflation were created. Current annualised issuance rate: 0.85% (negative = deflationary). Total supply: Ξ121,728,733. Live numbers update daily from growthepie's ETH supply tracker, which marks The Merge explicitly on the chart so the issuance step-down is visible. What changed Three big things, all immediate: 1. New ETH issuance: down 87%. Mining rewards (13,000 ETH/day) were replaced with staking rewards (1,700 ETH/day). The protocol now issues new ETH only as compensation for validators securing the network, not as compensation for energy-intensive computation. 2. Energy consumption: down 99.95%. Pre-Merge Ethereum mining consumed roughly 78 TWh/year at peak — comparable to Chile or Austria. Post-Merge, the entire validator infrastructure uses on the order of 0.01 TWh/year. The environmental case for Ethereum changed completely. 3. Conditions for net-deflation created. EIP-1559 (active since August 2021) was already burning the base fee of every transaction. The Merge cut new issuance enough that, on busy network days, burn now routinely exceeds issuance — making ETH net-deflationary for the first time. See /answers/is-ethereum-deflationary. What didn't change Critically: - Transaction fees stayed the same. The Merge didn't lower gas costs. - Throughput stayed the same. The Merge didn't scale Ethereum — that's the rollup-centric roadmap. - Smart contracts stayed the same. Every existing contract continued working without modification. - Users and developers didn't need to do anything. Wallets, dApps, and integrations kept working transparently. The Merge was a consensus-layer change — replacing miners with validators. It didn't touch the execution layer that users interact with. Proof-of-Work vs Proof-of-Stake — quick comparison | Aspect | Pre-Merge (PoW) | Post-Merge (PoS) | |---|---|---| | Block producer | Miner who finds the right hash | Validator pseudorandomly selected | | Hardware needed | GPUs or ASICs | Standard server / consumer hardware | | Energy use | 78 TWh/year at peak | 0.01 TWh/year | | New ETH per day | 13,000 ETH | 1,700 ETH | | Validator entry cost | Buy mining hardware | Stake 32 ETH (or up to 2,048 post-Pectra) | | Slashing risk | None (just wasted electricity) | Yes — dishonest validators lose stake | | Block time | 13s average (variable) | 12s (fixed slots) | Why did Ethereum switch? Three reasons: - Energy. PoW mining was environmentally indefensible at Ethereum's scale. The optics and ESG case were unsustainable. - Economics. PoW rewards external hardware suppliers and electricity producers. PoS keeps the economic flywheel internal to ETH holders. - Scaling roadmap. Future Ethereum upgrades (data availability sharding via Dencun, the broader rollup-centric roadmap, PeerDAS via Fusaka) needed PoS as a foundation. PoW would have been a hard ceiling on what Ethereum could become. Staking after The Merge Validators stake 32 ETH (or, after Pectra in May 2025, up to 2,048 ETH per consolidated validator — see /answers/what-pectra-upgrade-changed) and run validator software 24/7. The protocol randomly selects validators to propose each block (every 12 seconds) and attest to other validators' blocks. Honest validators earn staking rewards (3–5% APY on the staked ETH today). Dishonest ones get slashed — losing part of their stake. As of 2026, 30M+ ETH is staked across hundreds of thousands of validators. Staking is permissionless — anyone with 32 ETH can run a validator, and pools/liquid staking tokens (stETH, rETH) let smaller holders participate. The seven-year journey The Merge was the culmination of work that started in 2014 — before Ethereum mainnet had even launched. The major milestones: - 2014 — Vitalik Buterin proposes PoS for Ethereum - 2017 — Casper FFG research begins in earnest - December 2020 — Beacon Chain (PoS consensus) launches alongside existing PoW chain - August 2021 — EIP-1559 (the burn) activates — sets up half of what makes The Merge transformative - September 15, 2022 — The Merge: execution moves off PoW and onto Beacon Chain PoS It's one of the largest live-system upgrades in software engineering history — switching the consensus mechanism of a $200B+ network without any downtime. What came after Shanghai/Capella (April 2023) — enabled withdrawals of staked ETH. Before this, stakers could only deposit, not withdraw. Dencun (March 2024) — introduced blobs (EIP-4844), the foundation of cheap L2 fees. Pectra (May 2025) — doubled blob capacity, introduced EIP-7702 smart accounts, raised the validator stake cap. Fusaka (December 2025) — tripled blob capacity, added PeerDAS, introduced the EIP-7918 blob fee floor. Each of these built on the foundation The Merge established. PoS was the prerequisite for the rollup-centric scaling roadmap that defines Ethereum today. Methodology and data sources Historical facts (Merge activation date, 87% issuance reduction, 99.95% energy reduction) come from the Ethereum Foundation's post-Merge analysis and publicly verifiable on-chain data. Live numbers — total supply, current annualised issuance rate, net 30-day change — come from growthepie's ETH supply tracker ( ), updated daily. For per-block burn detail and a live cumulative-burn counter, see ultrasound.money. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. The Merge is an Ethereum protocol event — supporter relationships don't affect the description. Full list of supporters: growthepie.com/donate. Cross-check this answer. ethereum.org's Merge page is the canonical Ethereum Foundation reference. ultrasound.money tracks live post-Merge supply dynamics. Etherscan and Beaconcha.in show live validator and staking data. ### FAQ ### What was The Merge and how did it change Ethereum? **The Merge** activated on **September 15th, 2022** and transitioned Ethereum from **Proof-of-Work (PoW)** to **Proof-of-Stake (PoS)** — replacing miners with validators who stake ETH to secure the network. The three biggest immediate changes: **new ETH issuance dropped ~87%** (from ~13,000 ETH/day of mining rewards to ~1,700 ETH/day of staking rewards), **energy consumption fell ~99.95%** (no more electricity-intensive PoW mining), and **the conditions for ETH deflation were created** — combined with EIP-1559's base-fee burn (already active since August 2021), the lower issuance meant network activity could now burn more ETH than the protocol creates. Current annualised issuance rate: **%**. ### When did The Merge happen? **September 15th, 2022**, at 06:42 UTC, at terminal total difficulty of 58,750,000,000,000,000,000,000 (the trigger used to detect the transition). It was the culmination of seven years of research and development, having been on Ethereum's roadmap since the project's earliest days. The mainnet beacon chain (Ethereum's PoS consensus layer) had launched in December 2020; The Merge connected it to the execution layer that users interact with. ### What was Ethereum like before The Merge? **Proof-of-Work, like Bitcoin.** Specialised mining hardware (GPUs and ASICs) competed to find a hash satisfying the network's difficulty target. The miner who found the hash got the block reward — newly-issued ETH plus all transaction fees in the block. Mining was energy-intensive (Ethereum mining consumed ~78 TWh/year at peak — roughly the electricity use of a mid-sized country), and ETH issuance was high (~13,000 ETH/day, ~5M ETH/year). ### What is Ethereum like after The Merge? **Proof-of-Stake.** Validators stake ETH (32 ETH minimum per validator, raised to optional 2,048 ETH in the Pectra upgrade) and are pseudorandomly selected to propose blocks. Honest validators earn staking rewards proportional to their stake; dishonest validators get **slashed** (lose part of their stake). No mining hardware, ~99.95% less energy consumption, ~87% less new ETH issuance, and a foundation for further protocol upgrades (sharding, faster slot times) that PoW couldn't support. ### Why did Ethereum switch to Proof-of-Stake? Three main reasons. **(1) Energy.** PoW mining is environmentally indefensible at scale — Ethereum at peak used more electricity than several countries. **(2) Economics.** PoW rewards external hardware suppliers and energy producers; PoS rewards ETH holders directly, keeping the economic flywheel inside the network. **(3) Scaling.** PoS is a prerequisite for further protocol upgrades like data availability sharding (which came with Dencun in 2024) and the broader rollup-centric roadmap. PoW would have been a hard ceiling on Ethereum's evolution. ### How did The Merge affect ETH issuance? New ETH issuance dropped **~87% overnight** — from ~13,000 ETH/day of mining rewards to ~1,700 ETH/day of staking rewards. Combined with **EIP-1559** (which had already been burning the base fee of every transaction since August 2021), this created the conditions for **net-deflationary ETH**: even modest network activity could now burn more ETH than the protocol issues. Current annualised issuance rate (live): **%**. Net change in ETH supply over the last 30 days: **Ξ**. See [/answers/is-ethereum-deflationary](/answers/is-ethereum-deflationary). ### Is Ethereum deflationary because of The Merge? **The Merge created the conditions but didn't guarantee deflation.** It cut issuance by 87% but didn't add any burn mechanism — that was already there from EIP-1559 (August 2021). The combination is what makes net-deflation possible: low issuance + active burn. Whether ETH is net-deflationary on any given day still depends on network activity. High activity (DeFi, peak L1 usage) → more burn → deflationary. Low activity → less burn → mildly inflationary. As of right now the annualised rate is **%** (negative = deflationary, positive = inflationary). ### How much energy did The Merge save? Approximately **99.95% reduction** in Ethereum's electricity consumption. Pre-Merge Ethereum mining used roughly 78 TWh/year at peak — comparable to the electricity consumption of countries like Chile or Austria. Post-Merge, Ethereum's entire validator infrastructure uses on the order of 0.01 TWh/year. The environmental case for Ethereum changed completely with The Merge: from "one of the largest electricity consumers on Earth" to "energy footprint of a mid-sized data centre". ### What did The Merge NOT change? Critically: **transaction fees, transaction throughput, and existing smart contracts all stayed the same.** The Merge was purely a consensus-layer change — switching from miners to validators. It did **not** scale Ethereum (that's what L2s and the rollup-centric roadmap do); it did **not** lower gas fees directly (that's what Dencun and Fusaka eventually did via blob scaling for L2s); and it did **not** require any user or developer action. Existing wallets, contracts, and applications continued working without modification. ### How does staking work after The Merge? Validators **stake 32 ETH** (or, after Pectra in May 2025, up to 2,048 ETH per consolidated validator) and run validator software 24/7. The protocol randomly selects validators to **propose** each block (every 12 seconds) and to **attest** to other validators' blocks. Honest behaviour earns staking rewards (currently ~3–5% APY on the staked ETH); dishonest behaviour — equivocating on attestations or going offline for extended periods — gets **slashed**, removing part of the validator's stake. As of 2026, ~30M+ ETH is staked across hundreds of thousands of validators. ### Can anyone become an Ethereum validator? **Yes** — staking is permissionless. You need 32 ETH minimum (or you can join a staking pool with less, or use a liquid staking token like stETH or rETH). You also need to run validator software reliably; downtime gets penalised. Pectra (May 2025) added the option to run a single consolidated validator with up to 2,048 ETH staked — useful for institutions and large solo stakers but optional. See [/answers/what-pectra-upgrade-changed](/answers/what-pectra-upgrade-changed) for the Pectra staking-upgrade details. ### How long did The Merge take to develop? Roughly **seven years from research to mainnet**. Vitalik Buterin first proposed PoS for Ethereum in 2014 (before mainnet launched). The Casper FFG research program ran from 2017. The Beacon Chain (PoS consensus layer) launched in December 2020 and ran in parallel with the existing PoW chain for ~21 months. The Merge in September 2022 was the moment the two chains joined — execution moved off PoW and onto the Beacon Chain's PoS consensus. It's one of the largest live-system upgrades in software engineering history. ### What major upgrades came after The Merge? **Shanghai/Capella** (April 2023) — enabled staked-ETH withdrawals. **Dencun** (March 2024) — introduced blobs (EIP-4844), the foundation of cheap L2 fees. **Pectra** (May 2025) — doubled blob capacity, introduced EIP-7702 smart-account features, raised the validator effective stake cap to 2,048 ETH. See [/answers/what-pectra-upgrade-changed](/answers/what-pectra-upgrade-changed). **Fusaka** (December 2025) — tripled blob capacity, added PeerDAS, introduced the EIP-7918 blob fee floor. See [/answers/what-fusaka-upgrade-changed](/answers/what-fusaka-upgrade-changed). Each of these built on the foundation The Merge established. ### Where do these numbers come from? Historical facts (Merge activation date, 87% issuance reduction, 99.95% energy reduction) come from the Ethereum Foundation's post-Merge analysis and public on-chain data. Live numbers — current annualised issuance rate, net 30-day supply change — come from growthepie's ETH supply tracker (`/v1/eim/eth_supply.json`), which aggregates beacon chain data and burn events daily. For per-block burn detail and a cumulative-burn counter, see [ultrasound.money](https://ultrasound.money/). ### Where can I see live data on The Merge's effects? growthepie's [ETH supply tracker](/quick-bites/eth-supply) charts total supply and issuance rate over time, with **The Merge** explicitly marked as an annotation on the chart so you can see the issuance step-down visually. [ultrasound.money](https://ultrasound.money/) tracks live ETH burn and projected issuance. growthepie's [is-ethereum-deflationary answer](/answers/is-ethereum-deflationary) walks through the post-Merge supply dynamics in detail. --- ## Which Ethereum L2 has the most smart-account (EIP-7702) activity? URL: https://www.growthepie.com/answers/most-smart-account-activity-ethereum-l2 ### Direct answer Among Ethereum L2s, **Base** leads EIP-7702 (smart account / Type 4) transaction activity with 747.4k over the last 30 days, 4.37M all-time since Pectra. Full L2 ranking: 1. Base (747.4k over 30d; 4.37M all-time); 2. Arbitrum One (409.4k over 30d; 2.24M all-time); 3. OP Mainnet (120.1k over 30d; 329.7k all-time); 4. Unichain (6.6k over 30d; 52.9k all-time). For reference, Ethereum L1 sees 718.6k Type 4 transactions over the last 30 days (2.40M all-time). Data: 2026-05-27 UTC. Live tracker: growthepie.com/quick-bites/pectra-upgrade. ### Full answer Short answer (data 2026-05-27 UTC): Among Ethereum L2s, Base (747.4k over 30d; 4.37M all-time) leads EIP-7702 (smart-account / Type 4 transaction) activity. Full L2 ranking: 1. Base (747.4k over 30d; 4.37M all-time); 2. Arbitrum One (409.4k over 30d; 2.24M all-time); 3. OP Mainnet (120.1k over 30d; 329.7k all-time); 4. Unichain (6.6k over 30d; 52.9k all-time). For comparison, Ethereum L1: Ethereum L1 (718.6k over 30d; 2.40M all-time). EIP-7702 introduces the 'Set Code' / Type 4 transaction, activated with the Pectra upgrade on May 7, 2025. It enables smart-account features (gas sponsorship, paying gas in non-ETH tokens, transaction batching) for ordinary wallets without requiring users to switch to a dedicated smart-contract wallet. L2 ranking by EIP-7702 activity As of 2026-05-27 UTC, Base leads Ethereum L2s in EIP-7702 (smart-account / Type 4) transactions with 747.4k over the last 30 days (4.37M all-time since Pectra activated in May 2025). For comparison, Ethereum L1 sees 718.6k Type 4 transactions over 30 days. What EIP-7702 actually does EIP-7702 lets a regular Ethereum wallet (an EOA — externally-owned account) temporarily act as a smart contract for one transaction. The mechanism: the user signs an authorization that tells the protocol to "act as if I had this contract bytecode for this transaction". The chain installs the code temporarily; the transaction executes with the contract's behaviour; afterwards the EOA goes back to having no code. What this unlocks: - Gas sponsorship. A dApp, paymaster, or third party pays the gas fee on the user's behalf. - Pay gas in non-ETH tokens. Pay fees in USDC, a project token, or anything the wallet supports. - Transaction batching. Multiple actions in one user-signed transaction (no more + two-step). - Session keys / delegation. A dApp signs on the user's behalf within preset limits. Users don't need to switch wallets — their existing EOA opts in by signing an authorization. Adoption is gated by wallet support. How adoption is measured The simplest live signal is the number of Type 4 transactions per chain per day. A Type 4 transaction is a transaction whose field equals — these are EIP-7702 Set Code transactions by definition. growthepie tracks this per-chain in the Pectra fork data endpoint and exposes it on the Pectra upgrade tracker. High Type 4 counts mean wallets and dApps are actually using the upgrade. Low counts mean adoption is lagging despite the protocol feature being live. Why does L2 EIP-7702 activity matter? L2s are where most user-facing Ethereum activity happens today (see /answers/ethereum-l2-transaction-count). If EIP-7702 features genuinely improve UX, you'd expect adoption to be highest on L2s — where users are most price-sensitive and most likely to interact with consumer dApps that benefit from gas sponsorship. The L2 ranking above is the live read on whether that's happening. Methodology and data sources Per-chain Type 4 transaction counts come from growthepie's Pectra tracker endpoint ( , path ). Counts are derived from on-chain transaction-type analysis: a transaction is a Type 4 if its field is . The series starts on May 7, 2025 (Pectra activation) and updates daily. Currently growthepie's Pectra tracker publishes per-chain Type 4 counts for Ethereum L1, Base, OP Mainnet, Unichain, and Arbitrum — the chains where adoption has been most measurable. Coverage expands as more chains accumulate meaningful Type 4 activity. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Type 4 counts are computed mechanically from on-chain data — chains don't influence the ranking. Full list of supporters: growthepie.com/donate. Cross-check this answer. Etherscan lets you filter transactions by type — search Type 4 transactions per chain to verify counts. Per-chain block explorers (Basescan, Arbiscan, etc.) expose the same filter. growthepie's Pectra tracker has the live charts. Related answers - /answers/what-pectra-upgrade-changed — the full Pectra upgrade including EIP-7702, EIP-7691 (more blobs), and EIP-7251 (staking). - /answers/zk-vs-optimistic-rollup — the L2 architecture types EIP-7702 runs on top of. ### FAQ ### Which Ethereum L2 has the most smart-account (EIP-7702) activity? As of 2026-05-27 UTC, **Base (747.4k over 30d; 4.37M all-time)** leads Ethereum L2s in EIP-7702 (smart account / Type 4 transaction) activity. Full L2 ranking: 1. Base (747.4k over 30d; 4.37M all-time); 2. Arbitrum One (409.4k over 30d; 2.24M all-time); 3. OP Mainnet (120.1k over 30d; 329.7k all-time); 4. Unichain (6.6k over 30d; 52.9k all-time). Ethereum L1 (for context): **Ethereum L1 (718.6k over 30d; 2.40M all-time)**. EIP-7702 activated with Pectra on May 7, 2025. Live tracker: [growthepie.com/quick-bites/pectra-upgrade](/quick-bites/pectra-upgrade). ### What is EIP-7702? **EIP-7702** is a Pectra-era Ethereum upgrade (activated May 7, 2025) that introduces a new **"Set Code" or Type 4 transaction**. It lets a regular wallet (an EOA — externally-owned account) temporarily act as a smart contract account for one transaction. The unlocks: **gas sponsorship** (someone else pays the fee), **paying gas in tokens other than ETH** (USDC, project tokens), **transaction batching** (multiple actions in one user signature), and **session keys / delegation** (a dApp can sign on the user's behalf within preset limits). See [/answers/what-pectra-upgrade-changed](/answers/what-pectra-upgrade-changed) for the full Pectra story. ### What's a "Type 4" transaction? A new Ethereum transaction format introduced by EIP-7702 that carries an **authorization list** — signed instructions telling the protocol to temporarily install bytecode on the sender's EOA. While the transaction executes, the EOA behaves as a smart contract account; after the transaction completes the original (empty) code is restored. The chain accepts the temporary "set code" only if the EOA owner signed the authorization. Type 4 transactions count is the simplest live measure of EIP-7702 adoption. ### Which L2s show meaningful EIP-7702 adoption? As of 2026-05-27 UTC, the L2s with measurable Type 4 transaction activity tracked by growthepie are listed in the table above. The leader is **Base (747.4k over 30d; 4.37M all-time)**. Adoption depends heavily on **wallet support** — a chain only sees Type 4 transactions if its users' wallets expose EIP-7702 features. Wallets that adopted early (Coinbase Wallet, MetaMask, Rabby) drove most of the initial activity; chains with those wallets in their default user flow have the highest counts. ### How does L2 EIP-7702 activity compare to Ethereum L1? As of 2026-05-27 UTC, Ethereum L1 shows Ethereum L1 (718.6k over 30d; 2.40M all-time). L2 vs L1 activity can go either way depending on the use case: **dApps doing gas sponsorship for onboarding** tend to settle on L2s for cheaper transactions, so L2 Type 4 counts can outpace L1. **Native ETH transfers using EIP-7702 features** (e.g. paying with USDC for gas) often happen on L1 where ETH itself lives. The breakdown shifts as more wallets ship support. ### Why does EIP-7702 adoption matter? EIP-7702 is the most user-visible feature of the Pectra upgrade — it's the change that lets Ethereum dApps offer the kind of UX users get on centralised platforms (no gas token needed, batched approvals, "log in" sessions). High Type 4 transaction counts mean wallets and dApps are actually using the upgrade, not just citing it in announcements. Low counts mean adoption is lagging despite the protocol-level feature being live. ### Which use cases drive EIP-7702 transactions? Several patterns. **Onboarding gas sponsorship** — dApps paying gas for first-time users (often via Paymaster contracts integrated with EIP-7702). **DeFi batching** — combining ERC-20 `approve` + DEX swap into one transaction. **Game and social dApps** — session keys that authorise a series of in-app actions without re-signing each one. **Multi-step bridge UX** — wallets bundling approvals + bridge transactions atomically. Each pattern produces Type 4 transactions; the chain counts on this page sum them all. ### Where does this data come from? Per-chain Type 4 transaction counts come from growthepie's Pectra tracker endpoint (`/v1/quick-bites/pectra-fork.json`) — specifically the `data.type4_tx_count.{chain}.daily` series. Counts are derived from on-chain transaction-type analysis: a transaction is a Type 4 if its `type` field is `0x04` (EIP-7702 Set Code transactions). The series starts on May 7, 2025 (Pectra activation) and updates daily. ### Why don't all L2s appear in this ranking? growthepie's Pectra tracker currently publishes Type 4 transaction counts for the most widely-used EVM chains where EIP-7702 has measurable adoption — Ethereum L1, Base, OP Mainnet, Unichain, and Arbitrum. Chains without published Type 4 counts may not have had any (low adoption) or may not yet be in growthepie's tracker coverage. As wallet support broadens and more chains see meaningful Type 4 activity, the coverage list will expand. ### Where can I see this data live? growthepie's [Pectra upgrade tracker](/quick-bites/pectra-upgrade) charts per-chain Type 4 transactions over time. The same page tracks blob count vs target (EIP-7691 adoption). For the full Pectra upgrade context — what EIP-7702 actually does, why it matters, how it relates to account abstraction — see [/answers/what-pectra-upgrade-changed](/answers/what-pectra-upgrade-changed). ### How is "Ethereum L2" defined here? An Ethereum Layer 2 is a chain that derives security from Ethereum by posting transaction data and/or state to Ethereum mainnet. See [/answers/l2-vs-sidechain](/answers/l2-vs-sidechain) for the full definition. The L2s currently tracked in growthepie's Pectra Type 4 data are Base, OP Mainnet, Unichain, and Arbitrum. --- ## Where does most DeFi happen — Ethereum mainnet or L2s? URL: https://www.growthepie.com/answers/defi-l1-vs-l2 ### Direct answer DeFi activity has moved to L2s, but DeFi capital still concentrates on L1. Over the last 30 days, L2s processed 307.13M DeFi transactions (98% of L1+L2 total) vs 6.37M on Ethereum L1. By gas fees paid: L2s $13.49M (59%), L1 $9.38M. By stablecoin liquidity (the dominant pool of DeFi capital), L1 holds $176.11B vs $14.87B across all L2s (L2 share 7.8%). On per-transaction cost, a live swap on L1 costs 12¢ vs 0.020¢ on the cheapest L2 (599× more expensive on L1). Data: 2026-05-27 UTC. Live per-app DeFi rankings: growthepie.com/applications. ### Full answer Short answer (data 2026-05-27 UTC): Most DeFi activity happens on L2s, but DeFi capital still concentrates on L1. Over the last 30 days, L2s processed 307.13M DeFi transactions vs 6.37M on Ethereum L1 — L2 share 98% by transactions, 59% by gas fees paid. By stablecoin liquidity, L1 holds $176.11B vs $14.87B across L2s (L2 share 7.8%). Top L2 DeFi contributors: 1. Base Chain (117.70M txs, 64% of L2 DeFi); 2. OP Mainnet (34.50M txs, 19% of L2 DeFi); 3. Arbitrum One (14.18M txs, 7.8% of L2 DeFi); 4. MegaETH (7.18M txs, 3.9% of L2 DeFi); 5. Starknet (5.37M txs, 2.9% of L2 DeFi). Sourced from for the per-chain DeFi (finance-category) activity, plus per-chain stablesmcap endpoints for the capital comparison and the live fees table for the swap-fee comparison. L1 vs L2 DeFi at a glance As of 2026-05-27 UTC, Over the last 30 days, L2s processed 307.13M DeFi transactions (98% of L1+L2 total), vs 6.37M on Ethereum L1. By gas fees paid: L2s $13.49M (59%), L1 $9.38M (41%). By stablecoin liquidity (the dominant pool of DeFi capital), L1 holds $176.11B vs $14.87B across all L2s (L2 share: 7.8%). Why has DeFi activity moved to L2s? Three forces: - Fee economics. DeFi transactions are gas-heavy — even a "cheap" swap consumes 100k+ gas, plus the cost of approvals, slippage checks, and routing. Live (latest hour, data 2026-05-27 UTC): Ethereum L1: 12¢; cheapest L2 (Unichain): 0.020¢ — L1 currently costs 599× more than the cheapest L2. For active traders the difference is decisive. Pulled from the same feed that powers /answers/lowest-fee-ethereum-l2. - Composability within L2 ecosystems. Major L2s (Arbitrum, Base, OP Mainnet) host their own deep DeFi ecosystems — native AMMs, lending markets, perps DEXes. Cross-protocol DeFi composability is now practical without leaving the chain. - Network effects. Once liquidity concentrates on a chain, more protocols deploy there to access it, attracting more liquidity. L2 DeFi is now self-sustaining. Does this mean L1 DeFi is dead? No — far from it. Ethereum L1 holds $176.11B in stablecoins vs $14.87B across all L2s (live, data 2026-05-27 UTC). Stablecoins are the single biggest pool of DeFi capital, so this is the cleanest "where does the money sit" signal — and it still leans heavily L1. Major DeFi blue chips remain on L1 because of legacy liquidity, integration with high-value institutional positions, and the security profile of Ethereum mainnet. Maker/Sky, Aave's main markets, Curve's reference markets, and large Uniswap V3 pools all live on L1. The pattern: L2s dominate by transaction count and user activity; L1 leads in stablecoin liquidity and per-transaction value. Big trades and treasury operations still settle through L1 for the security guarantees; retail-scale activity has migrated to L2s for the cost savings. Where to see per-app DeFi rankings The largest L2 DeFi ecosystems by activity in 2026 are Arbitrum One, Base, OP Mainnet, and Unichain — these chains host the major DEX deployments (Uniswap, PancakeSwap, Aerodrome) and lending protocols (Aave, Morpho, Spark) that drive most L2 DeFi activity. For live per-app DeFi rankings on L2s (transaction counts, gas fees, active addresses per protocol) see growthepie.com/applications and /answers/top-apps-ethereum-l2s. Methodology and data sources We deliberately use only small, cacheable endpoints so this page renders fast under Next.js SSR (the 2MB fetch-cache ceiling rules out some otherwise-useful sources — see methodology FAQ). 1. L1 vs L2 capital (stablecoin supply) — per-chain stablesmcap endpoints. for L1 and for the L2 aggregate (with a per-chain sum fallback). Latest daily USD value. Stablecoins are the single biggest pool of DeFi capital, so this is the cleanest "where does DeFi money sit" proxy. We use it instead of TVL because per-chain TVL isn't exposed for Ethereum L1 in growthepie's API today. 2. Live swap-fee comparison — , path (USD column). Picks Ethereum L1's value and the minimum across the L2 universe. Why no per-window L1 vs L2 DeFi transaction count on this page? growthepie's blockspace category endpoint ( ) does have per-L2 DeFi transactions, but the file is 55MB — above Next.js's 2MB SSR fetch-cache ceiling, so it would re-download on every render (multi-minute page loads). Per-app DeFi rankings (which use the same data, queried client-side) live at growthepie.com/applications. For L1 DeFi specifically, DefiLlama is the best source — growthepie doesn't break Ethereum L1 transactions into DeFi/non-DeFi categories. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Numbers on this page are computed mechanically from public API data. Full list of supporters: growthepie.com/donate. Cross-check this answer. DefiLlama tracks DeFi TVL split between L1 and L2s (independent verification of the stablecoin-supply ratio). L2BEAT's activity view tracks per-L2 activity. growthepie's applications page is the live per-app DeFi ranking. Related answers - /answers/is-ethereum-scaling-through-l2s — the broader L1 vs L2 activity comparison (all categories, not just DeFi). - /answers/top-apps-ethereum-l2s — per-app ranking on L2s including the major DeFi apps. - /answers/most-value-secured-ethereum-l2 — TVS ranking, the "where does the capital sit" question that's adjacent to "where does DeFi happen". ### FAQ ### Where does most DeFi happen — Ethereum mainnet or L2s? **Most DeFi activity happens on L2s, but DeFi capital still concentrates on L1.** Over the last 30 days (data 2026-05-27 UTC), Ethereum L2s processed **307.13M** DeFi transactions vs **6.37M** on Ethereum L1 — L2s account for **98%** of all L1+L2 DeFi activity by transaction count, **59%** by gas fees paid. By **stablecoin liquidity** (the dominant pool of DeFi capital), L1 holds **$176.11B** vs **$14.87B** across all L2s — L2 share **7.8%**. Top L2 DeFi contributors: 1. Base Chain (117.70M txs, 64% of L2 DeFi); 2. OP Mainnet (34.50M txs, 19% of L2 DeFi); 3. Arbitrum One (14.18M txs, 7.8% of L2 DeFi); 4. MegaETH (7.18M txs, 3.9% of L2 DeFi); 5. Starknet (5.37M txs, 2.9% of L2 DeFi). Live: [growthepie.com/blockspace/chain-overview](https://www.growthepie.com/blockspace/chain-overview). ### What counts as "DeFi" here? growthepie's blockspace classification groups DeFi into four subcategories: **DEX** (decentralised exchanges), **lending** (Aave, Compound, etc.), **derivative** (perps DEXes, options, structured products), and **staking** (liquid staking like Lido, Rocket Pool, plus restaking platforms). A transaction is "DeFi" if it touches a labelled contract in any of these subcategories. The classification is based on labelled contract data in growthepie's labels system — see [growthepie.com/applications](https://www.growthepie.com/applications) for the full per-app classification. ### Why has DeFi moved to L2s? Three forces. **(1) Fee economics** — DeFi transactions are gas-heavy (swaps, lending operations, vault deposits often consume 100k+ gas). At Ethereum L1 fees, even a "cheap" swap can cost $5–50. On L2s the same swap costs cents. **(2) Composability within L2 ecosystems** — major L2s (Arbitrum, Base, OP Mainnet) have their own deep DeFi ecosystems with native protocols, making cross-protocol composability practical without leaving the chain. **(3) Network effects** — once liquidity is on a chain, more protocols deploy there to access that liquidity, attracting more liquidity. L2 DeFi is now self-sustaining. ### Does this mean Ethereum L1 DeFi is dead? No. Ethereum L1 still hosts the **largest absolute TVL** of any chain — major DeFi blue chips (Maker/Sky, Aave's main markets, Curve's reference markets, large Uniswap pools) remain on L1 because of legacy liquidity, integration with high-value institutional positions, and the security profile of mainnet. **By transaction count L2s dominate; by per-transaction value L1 often leads.** Big trades and treasury operations still go through L1; retail-scale activity has migrated to L2s. ### How much L2 DeFi activity is there across different time windows? We don't surface live per-window L2 DeFi transaction counts on this page because growthepie's blockspace category data file (`/v1/blockspace/category_comparison.json`) is ~55MB — beyond Next.js's 2MB fetch-cache ceiling, so reading it from SSR would re-download the file every render (multi-minute page loads). For per-app DeFi rankings on L2s with live transaction counts, see [growthepie.com/applications](https://www.growthepie.com/applications) — the same data, queried client-side where the cache size limit doesn't apply. ### Which L2 has the most DeFi activity? The largest L2 DeFi ecosystems in 2026 are Arbitrum One, Base, OP Mainnet, and Unichain — they host the major DEX deployments (Uniswap, PancakeSwap, Aerodrome) and lending protocols (Aave, Morpho, Spark) that drive the bulk of L2 DeFi activity. For live per-app rankings (with transaction counts and gas fees) see [growthepie.com/applications](https://www.growthepie.com/applications) — that page reads the same data client-side, where Next.js's 2MB SSR cache ceiling doesn't apply. ### What are the main DeFi subcategories? growthepie's DeFi category breaks into four subcategories: **DEX** (swap, AMM, orderbook trading), **Lending** (overcollateralised lending markets), **Derivative** (perpetuals, options, structured products), and **Staking** (liquid staking, restaking, validator pools). The four together make up "DeFi" on this page. For per-subcategory breakdown see growthepie's [blockspace category comparison](https://www.growthepie.com/blockspace/category-comparison). ### How is this calculated? growthepie's [blockspace overview endpoint](https://api.growthepie.com/v1/blockspace/overview.json) exposes per-chain per-category activity rollups at multiple window lengths (7d / 30d / 90d / 365d / max). For Ethereum L1 we read `chains.ethereum.overview[window].defi.data`; for L2 ecosystem we read `chains.all_l2s.overview[window].defi.data`. Each `.defi.data` row carries transactions and USD gas-fees-paid in the DeFi category. We compute L1+L2 totals and the L2 share from those two rows. ### Why does this use the `all_l2s` aggregate? Because the question is "L1 vs L2 collectively", not "L1 vs each individual L2". growthepie's `all_l2s` aggregate sums activity across every L2 it tracks — much cleaner than asking us to sum manually. For the per-L2 breakdown, see the second table on this page (top 10 L2s by DeFi transactions). For activity rankings across all metrics (not just DeFi), see [/answers/most-used-ethereum-l2](/answers/most-used-ethereum-l2). ### How does this compare to "is Ethereum scaling through L2s"? [/answers/is-ethereum-scaling-through-l2s](/answers/is-ethereum-scaling-through-l2s) compares L1 vs L2 across **all** transaction activity. This page narrows it to **DeFi only**. The DeFi share on L2s tends to be higher than the all-activity share because DeFi transactions are particularly fee-sensitive and benefit most from L2 cost savings. ### Is Polygon PoS counted as an L2 here? No. Polygon PoS is a sidechain — it doesn't belong in the L2 ecosystem aggregate. See [/answers/l2-vs-sidechain](/answers/l2-vs-sidechain). The `all_l2s` aggregate growthepie uses follows the same exclusion list. ### Where can I see this live? growthepie's [blockspace chain overview](https://www.growthepie.com/blockspace/chain-overview) and [blockspace category comparison](https://www.growthepie.com/blockspace/category-comparison) charts both expose this data interactively. Filter to the "defi" category to see per-chain per-window splits. --- ## Which Ethereum L2 has the lowest fees? URL: https://www.growthepie.com/answers/lowest-fee-ethereum-l2 ### Direct answer By median transaction fee, the daily-cheapest Ethereum L2 is Manta Pacific (0.000414¢); weekly-cheapest: Manta Pacific (0.000478¢); monthly-cheapest: Manta Pacific (0.000629¢). Live (latest hour) — median fee: Mode Network (0.0100¢); native transfer fee: Mode Network (0.0100¢); token swap fee: Unichain (0.0200¢); average fee: Mode Network (0.0100¢). Data: 2026-05-27 UTC. Live leaderboards: growthepie.com/fees. ### Full answer Short answer (data 2026-05-27 UTC): By median transaction fee the daily-cheapest Ethereum L2 is Manta Pacific (0.000414¢). Live (latest hour) — native transfer cheapest: Mode Network (0.0100¢). Token swap cheapest: Unichain (0.0200¢). Average fee cheapest: Mode Network (0.0100¢). Weekly and monthly median leaders may differ — see the per-period sections below. Updated daily — every leaderboard on this page is recomputed from growthepie's public API. Live fee values use the last completed hour; daily / weekly / monthly median values use the most recent completed period (not the in-progress one). How we measure "lowest fees" We rank Ethereum L2s by four fee facets, two of which are exposed at both live and historical granularities: - Median fee — cost of a typical transaction on the chain. Most quotable single number; available live (hourly) and historical (daily / weekly / monthly). - Native transfer fee — median cost of sending the chain's gas token. The cheapest common action and a useful floor. Live only. - Token swap fee — median cost of a representative DEX swap. The most common DeFi action; usually the most expensive of the four categories because swaps touch multiple smart contracts. Live only. - Average fee — arithmetic mean across all transactions on the chain. Can run higher than the median when a single block contains a few expensive transactions. Live only. Live leaderboard: by fee type (latest hour) Live L2 fee leaders (latest hour, 2026-05-27 UTC) — median fee: Mode Network (0.0100¢); native transfer fee: Mode Network (0.0100¢); token swap fee: Unichain (0.0200¢); average fee: Mode Network (0.0100¢). Live leaderboard: median fee Top 3 cheapest right now (latest completed hour, 2026-05-27 UTC): 1. Mode Network (0.0100¢), 2. Unichain (0.0100¢), 3. Base Chain (0.190¢). Live leaderboard: native transfer fee Top 3 cheapest right now (latest completed hour, 2026-05-27 UTC): 1. Mode Network (0.0100¢), 2. Base Chain (0.0300¢), 3. Celo (0.0300¢). Live leaderboard: token swap fee Top 3 cheapest right now (latest completed hour, 2026-05-27 UTC): 1. Unichain (0.0200¢), 2. Mode Network (0.0400¢), 3. Celo (0.300¢). Live leaderboard: average fee Top 3 cheapest right now (latest completed hour, 2026-05-27 UTC): 1. Mode Network (0.0100¢), 2. Unichain (0.0200¢), 3. OP Mainnet (0.100¢). Historical leaderboard: median fee (daily / weekly / monthly) As of 2026-05-27 UTC, Manta Pacific is the cheapest Ethereum L2 by median fee at 0.000414¢ daily, 0.000478¢ weekly, 0.000629¢ monthly. Daily top 3: 1. Manta Pacific (0.000414¢), 2. OP Mainnet (0.00222¢), 3. Mode Network (0.00392¢). Weekly top 3: 1. Manta Pacific (0.000478¢), 2. OP Mainnet (0.00142¢), 3. Soneium (0.00428¢). Monthly top 3: 1. Manta Pacific (0.000629¢), 2. OP Mainnet (0.00402¢), 3. Ink (0.00655¢). Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue and filter to chains where , , and the chain key is not on the explicit non-L2 list below. 2. For live fees, pull and read the latest completed hour from each of the four fee fields ( , , , ), USD column. 3. For historical median fee, pull the per-chain endpoint ( ) — the API exposes daily / weekly / monthly aggregations natively. Daily uses the last completed day; weekly and monthly use the most recent completed period. 4. Sort the chains ascending (cheapest first) for each (fee type, window) pair and take the top 3. All rankings on this page pull live from growthepie's public API and refresh daily; the values shown above were generated on 2026-05-27 UTC: - Master chain list (with bucket / chaintype classification): - Live hourly fee table (median / transfer / swap / average per chain): - Per-chain median fee timeseries (daily / weekly / monthly, USD): Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Some supporters operate chains that appear in the rankings above. Rankings are computed mechanically from public API data — chains do not pay for inclusion or placement, and supporters do not receive ranking adjustments or preferential treatment. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. Independent L2 fee data sources you can compare against include L2BEAT (activity, stage classification, risk analysis) and the chains' own block explorers. Methodologies and chain inclusion lists differ between providers — when rankings disagree, comparing the underlying fee definitions (which transactions are sampled? what time window?) is usually more informative than the ranks themselves. Which chains are included? While growthepie tracks the fees of every L2 in the Ethereum ecosystem, our detailed leaderboards focus on the most widely used and adopted chains. The list of 25 chains is computed automatically from and refreshed when growthepie adds or removes coverage: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Ethereum mainnet — it is Layer 1, not Layer 2. - Polygon PoS — a sidechain with its own validator set, not a Layer 2. - Aggregate keys ( , ) — not individual chains. ### FAQ ### Which Ethereum L2 has the lowest fees? It depends on the fee type and the time horizon. As of 2026-05-27 UTC, by median transaction fee the daily-cheapest L2 is Manta Pacific (0.000414¢). Live (latest hour): the cheapest L2 for native transfers is Mode Network (0.0100¢); for token swaps: Unichain (0.0200¢); by average fee: Mode Network (0.0100¢). Weekly and monthly leaders may differ — see the per-period FAQ below. All rankings are computed daily from growthepie's public API; live fee table: growthepie.com/fees. ### Why does the answer depend on the fee type? Different fee types measure different onchain actions. Median fee is the cost of a typical transaction on the chain — robust to outliers and the most quotable single number. Native transfer fee is the median cost of sending ETH (or the chain's gas token) — the cheapest common action and a good lower-bound benchmark. Swap fee is the median cost of a token swap — the most common DeFi action and the one users feel most often. Average fee is the mean fee, which can be skewed by a small number of expensive transactions and so reads higher than the median. ### Which Ethereum L2 has the cheapest median fee right now? Live (latest completed hour, data 2026-05-27 UTC), the top three Ethereum L2s by lowest median transaction fee are 1. Mode Network (0.0100¢), 2. Unichain (0.0100¢), 3. Base Chain (0.190¢). Live fee table: growthepie.com/fees. ### Which Ethereum L2 has the cheapest native transfer right now? Live (latest completed hour, data 2026-05-27 UTC), the top three Ethereum L2s by lowest native transfer fee are 1. Mode Network (0.0100¢), 2. Base Chain (0.0300¢), 3. Celo (0.0300¢). Native transfer fee is the median cost of sending the chain's gas token — the cheapest common transaction type, useful as a floor. ### Which Ethereum L2 has the cheapest token swap right now? Live (latest completed hour, data 2026-05-27 UTC), the top three Ethereum L2s by lowest token swap fee are 1. Unichain (0.0200¢), 2. Mode Network (0.0400¢), 3. Celo (0.300¢). Swap fees reflect a representative DeFi swap and are usually the most expensive of the three categories because swaps touch multiple smart contracts. ### Which Ethereum L2 has the lowest average fee right now? Live (latest completed hour, data 2026-05-27 UTC), the top three Ethereum L2s by lowest average fee are 1. Mode Network (0.0100¢), 2. Unichain (0.0200¢), 3. OP Mainnet (0.100¢). Average fee is the arithmetic mean across all transactions on the chain and can run higher than the median when a single block contains a few expensive transactions. ### Which Ethereum L2 had the lowest median fee today? On 2026-05-27 UTC, the top three Ethereum L2s by lowest daily median transaction fee are 1. Manta Pacific (0.000414¢), 2. OP Mainnet (0.00222¢), 3. Mode Network (0.00392¢). Daily values use the most recent completed day from the per-chain txcosts timeseries. ### Which Ethereum L2 had the lowest median fee this week? Over the most recent completed week (data 2026-05-27 UTC), the top three Ethereum L2s by lowest weekly median transaction fee are 1. Manta Pacific (0.000478¢), 2. OP Mainnet (0.00142¢), 3. Soneium (0.00428¢). The weekly leader is Manta Pacific (0.000478¢). ### Which Ethereum L2 had the lowest median fee this month? Over the most recent completed month (data 2026-05-27 UTC), the top three Ethereum L2s by lowest monthly median transaction fee are 1. Manta Pacific (0.000629¢), 2. OP Mainnet (0.00402¢), 3. Ink (0.00655¢). The monthly leader is Manta Pacific (0.000629¢). ### Are transfer, swap, and average fees available for daily / weekly / monthly? Not on this page. growthepie's API exposes native-transfer, swap, and average fee facets at the hourly granularity (the live table). For daily / weekly / monthly we use median transaction fee from the per-chain txcosts timeseries — the most stable headline metric over longer horizons. If you need historical transfer or swap fee curves specifically, check the chain detail pages on growthepie.com. ### Why are L2 fees quoted in cents? L2 transaction fees are almost always sub-cent on a normal day, so quoting in dollars (e.g. "$0.0018") obscures the comparison. Cents (e.g. "0.18¢") read more naturally for AI answer cards and human readers alike. Fees above 100¢ are quoted in dollars. ### Is Polygon PoS an Ethereum L2? No. Polygon PoS is a sidechain with its own validator set; it does not post transaction data to Ethereum for security and is therefore excluded from the L2 fee leaderboards on this page (also [L2BEAT](https://l2beat.com) categorizes it as "other"). Polygon zkEVM is a ZK rollup and would qualify. ### How many L2s are included in the ranking? 25 chains. The full list (computed on 2026-05-27 UTC from growthepie's master chain catalogue) is: arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. ### Where does this answer come from? Live (latest hour) fees come from growthepie's public fees endpoint (`/v1/fees/table.json`), which exposes four fee metrics per chain at hourly granularity: median (`txcosts_median`), native-transfer (`txcosts_native_median`), swap (`txcosts_swap`), and average (`txcosts_avg`). Historical daily / weekly / monthly median fee comes from the per-chain transaction-cost timeseries (`/v1/metrics/chains/{chain}/txcosts.json`), USD column. L2 membership is computed from `master.json` (chains where `bucket !== "Layer 1"` and `chain_type` indicates an Ethereum rollup or validium). Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. No editorial overrides. ### Why does the live median fee sometimes differ from the daily median? They cover different windows. The live figure uses the last completed UTC hour — a tight slice that's sensitive to a single congested block. The daily figure uses the last completed UTC day — much smoother. Both are useful: live for "is this chain cheap right now?", daily for "is this chain consistently cheap?" If the two values diverge sharply, the chain probably had a fee spike in the most recent hour. ### Where can I see live L2 fee data? growthepie's dedicated fees view at [fees.growthepie.com](https://fees.growthepie.com) tracks live (hourly) median, native-transfer, and swap fees per chain, plus throughput and TPS. The same data backs this answer page; the dashboard is interactive and adds chart overlays. ### How is "Ethereum L2" defined here? An Ethereum Layer 2 is a chain that derives security from Ethereum by posting transaction data and/or state to Ethereum mainnet. This includes optimistic rollups, ZK rollups, and Validiums. Sidechains (independent validator sets, like Polygon PoS) are excluded. --- ## Which Ethereum L2 has the most stablecoin activity? URL: https://www.growthepie.com/answers/most-stablecoin-activity-ethereum-l2 ### Direct answer By stablecoin supply, the daily leader is Arbitrum One ($8.02B); weekly leader: Arbitrum One ($7.90B); monthly leader: Arbitrum One ($7.54B). By stablecoin transactions, the daily leader is Celo (590.8k); weekly leader: Celo (5.88M); monthly leader: Celo (27.45M). By gas spent on stablecoins, the daily leader is Celo ($2.4k); weekly leader: Celo ($20.3k); monthly leader: Celo ($99.4k). By distinct stablecoin variety, Base Chain hosts the most (52 tokens). Data: 2026-05-27 UTC. Live leaderboards: growthepie.com/fundamentals/stablecoin-market-cap. ### Full answer Short answer (data 2026-05-27 UTC): By stablecoin supply the leader is Arbitrum One ($8.02B). By daily stablecoin transactions: Celo (590.8k). By gas spent on stablecoins: Celo ($2.4k). By distinct stablecoin variety: Base Chain (52 stablecoins). Weekly and monthly leaders may differ — see the per-metric leaderboards below. Updated daily — every leaderboard on this page is recomputed from growthepie's public API. Daily values use the latest available day; weekly and monthly use the most recent completed period (or for blockspace metrics, the latest 7d / 30d rolling window). How we measure "stablecoin activity" We rank Ethereum L2s by four dimensions that capture different facets of stablecoin presence: - Stablecoin supply — total USD value of every stablecoin parked on the chain. A balance-sheet signal; the chain with the most stablecoins as a store of value. - Stablecoin transactions — count of stablecoin transfers per period. A velocity signal; how often the stablecoins on the chain actually move. - Gas spent on stablecoins — fee revenue (USD) the chain earned from stablecoin transactions. A monetization signal; how economically important stablecoins are to the chain. - Variety — number of distinct stablecoin tokens deployed on the chain. A breadth signal; how many issuers and currencies are represented (USDC, USDT, DAI, PYUSD, EURC, etc.). Live leaderboard: stablecoin supply As of 2026-05-27 UTC, Arbitrum One leads Ethereum L2 stablecoin supply at $8.02B daily, $7.90B weekly, $7.54B monthly. Daily top 3: 1. Arbitrum One ($8.02B), 2. Base Chain ($4.57B), 3. Mantle ($524.92M). Weekly top 3: 1. Arbitrum One ($7.90B), 2. Base Chain ($4.57B), 3. Mantle ($556.67M). Monthly top 3: 1. Arbitrum One ($7.54B), 2. Base Chain ($4.70B), 3. Mantle ($692.34M). Live leaderboard: stablecoin transactions As of 2026-05-27 UTC, Celo leads Ethereum L2 stablecoin transactions at 590.8k daily, 5.88M weekly, 27.45M monthly. Daily top 3: 1. Celo (590.8k), 2. Base Chain (384.8k), 3. Arbitrum One (168.8k). Weekly top 3: 1. Celo (5.88M), 2. Base Chain (3.05M), 3. Arbitrum One (1.25M). Monthly top 3: 1. Celo (27.45M), 2. Base Chain (12.56M), 3. Arbitrum One (5.89M). Live leaderboard: gas spent on stablecoins As of 2026-05-27 UTC, Celo leads Ethereum L2 gas spent on stablecoins at $2.4k daily, $20.3k weekly, $99.4k monthly. Daily top 3: 1. Celo ($2.4k), 2. Base Chain ($565.86), 3. Arbitrum One ($501.89). Weekly top 3: 1. Celo ($20.3k), 2. Base Chain ($4.7k), 3. Arbitrum One ($3.8k). Monthly top 3: 1. Celo ($99.4k), 2. Base Chain ($19.6k), 3. Arbitrum One ($18.5k). Live leaderboard: variety of stablecoins As of 2026-05-27 UTC, Base Chain hosts the most distinct stablecoins of any tracked Ethereum L2 with 52 tokens deployed. Top 3: 1. Base Chain (52 stablecoins), 2. Arbitrum One (43 stablecoins), 3. Celo (23 stablecoins). Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue and filter to chains where , , and the chain key is not on the explicit non-L2 list below. 2. For stablecoin supply, pull the per-chain endpoint ( ) — daily / weekly / monthly aggregations exposed natively, USD column. 3. For stablecoin transactions and gas spent, pull the blockspace category endpoint ( ) and read . Daily = latest row of the series. Weekly = . Monthly = . Fields: for transactions, for gas spent. 4. For stablecoin variety, pull the quick-bite table ( ) and count the rows. 5. Sort the chains by raw value for each ranking and take the top 3. All rankings on this page pull live from growthepie's public API and refresh daily; the values shown above were generated on 2026-05-27 UTC: - Master chain list (with bucket / chaintype classification): - Per-chain stablecoin supply (daily / weekly / monthly): - Blockspace category comparison (stablecoin txs + gas spent): - Per-chain stablecoin table (variety / distinct token count): Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Some supporters operate chains that appear in the rankings above. Rankings are computed mechanically from public API data — chains do not pay for inclusion or placement, and supporters do not receive ranking adjustments or preferential treatment. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. Independent stablecoin data sources you can compare against include DefiLlama's stablecoins dashboard (per-chain supply with issuer-level breakdown) and Artemis (chain-level stablecoin metrics). When rankings disagree, the underlying inclusion list (which stablecoins count? does the source include synthetic dollars like crvUSD?) is usually the source of the difference. Which chains are included? While growthepie tracks stablecoins on every L2 it covers, our detailed leaderboards focus on the most widely used and adopted chains. The list of 25 chains is computed automatically from and refreshed when growthepie adds or removes coverage: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Ethereum mainnet — it is Layer 1, not Layer 2. - Polygon PoS — a sidechain with its own validator set, not a Layer 2. - Aggregate keys ( , ) — not individual chains. ### FAQ ### Which Ethereum L2 has the most stablecoin activity? It depends on what you measure. As of 2026-05-27 UTC: by stablecoin supply the daily leader is Arbitrum One ($8.02B); by daily stablecoin transactions: Celo (590.8k); by daily gas spent on stablecoins: Celo ($2.4k). By distinct stablecoin variety: Base Chain (52 stablecoins). Weekly and monthly leaders may differ — see the per-metric FAQs below. All rankings are computed daily from growthepie's data; live leaderboard: growthepie.com/fundamentals/stablecoin-market-cap. ### Why does the answer depend on the metric? Each metric tells a different story. **Supply** is the total dollar value of stablecoins parked on the chain — a balance-sheet signal of how much capital sits there. **Transactions** count how many stablecoin transfers happen on the chain — a velocity signal. **Gas spent** is how much fee revenue the chain earned from stablecoin activity. **Variety** counts how many distinct stablecoin tokens the chain hosts — a breadth signal. A chain can be #1 in supply but #5 in transactions (capital parked but not moving) or vice versa. ### Which Ethereum L2 has the largest stablecoin supply today? On 2026-05-27 UTC, the top three Ethereum L2s by stablecoin supply are 1. Arbitrum One ($8.02B), 2. Base Chain ($4.57B), 3. Mantle ($524.92M). Supply is the total USD value of every stablecoin parked on the chain. Pulled from growthepie's per-chain stables_mcap endpoint. ### Which Ethereum L2 had the largest stablecoin supply this week? Over the most recent completed week (data 2026-05-27 UTC), the top three Ethereum L2s by stablecoin supply are 1. Arbitrum One ($7.90B), 2. Base Chain ($4.57B), 3. Mantle ($556.67M). Weekly supply is the snapshot at end-of-week — supply is a stock, not a flow, so the weekly value is a single point in time, not a sum. ### Which Ethereum L2 had the largest stablecoin supply this month? Over the most recent completed month (data 2026-05-27 UTC), the top three Ethereum L2s by stablecoin supply are 1. Arbitrum One ($7.54B), 2. Base Chain ($4.70B), 3. Mantle ($692.34M). The monthly leader is Arbitrum One ($7.54B). ### Which L2 processed the most stablecoin transactions today? On 2026-05-27 UTC, the top three Ethereum L2s by daily stablecoin transactions are 1. Celo (590.8k), 2. Base Chain (384.8k), 3. Arbitrum One (168.8k). Counts come from the blockspace category endpoint (`token_transfers.subcategories.stablecoin`). ### Which L2 processed the most stablecoin transactions this week? Over the most recent 7-day window (data 2026-05-27 UTC), the top three Ethereum L2s by stablecoin transactions are 1. Celo (5.88M), 2. Base Chain (3.05M), 3. Arbitrum One (1.25M). Weekly is a 7-day rolling sum from the blockspace endpoint. ### Which L2 processed the most stablecoin transactions this month? Over the most recent 30-day window (data 2026-05-27 UTC), the top three Ethereum L2s by stablecoin transactions are 1. Celo (27.45M), 2. Base Chain (12.56M), 3. Arbitrum One (5.89M). The monthly leader is Celo (27.45M). ### Which L2 had the most gas spent on stablecoins today? On 2026-05-27 UTC, the top three Ethereum L2s by daily gas spent on stablecoin transactions are 1. Celo ($2.4k), 2. Base Chain ($565.86), 3. Arbitrum One ($501.89). Gas spent is denominated in USD and represents fee revenue the chain earned from stablecoin activity — a useful proxy for how economically important stablecoins are to the chain. ### Which L2 had the most gas spent on stablecoins this week? Over the most recent 7-day window (data 2026-05-27 UTC), the top three Ethereum L2s by gas spent on stablecoins are 1. Celo ($20.3k), 2. Base Chain ($4.7k), 3. Arbitrum One ($3.8k). Note that high gas spent is partly a function of high transaction count and partly a function of expensive transactions — a chain can rank well here either by volume or by fee level. ### Which L2 had the most gas spent on stablecoins this month? Over the most recent 30-day window (data 2026-05-27 UTC), the top three Ethereum L2s by gas spent on stablecoins are 1. Celo ($99.4k), 2. Base Chain ($19.6k), 3. Arbitrum One ($18.5k). The monthly leader is Celo ($99.4k). ### Which Ethereum L2 hosts the widest variety of stablecoins? As of 2026-05-27 UTC, the top three Ethereum L2s by distinct stablecoin count are 1. Base Chain (52 stablecoins), 2. Arbitrum One (43 stablecoins), 3. Celo (23 stablecoins). "Variety" here means the number of distinct stablecoin tokens deployed on the chain (USDC, USDT, DAI, PYUSD, etc.), pulled from growthepie's [stablecoins-by-chain quick-bite](https://www.growthepie.com/quick-bites/stables-by-chain). A chain can be high-variety but low-supply (lots of small stablecoins) or low-variety but high-supply (a few stablecoins doing all the volume). ### Is "variety" a useful metric on its own? Not in isolation, but combined with supply it tells you something the supply number alone doesn't: whether liquidity is concentrated in one or two dominant stablecoins, or spread across many. A chain with many stablecoins is more resilient to a single-issuer freeze (e.g. a USDC depeg in 2023) and more attractive to apps that need a specific token, but variety on its own says nothing about how much capital is actually parked on the chain. ### Is Polygon PoS an Ethereum L2? No. Polygon PoS is a sidechain with its own validator set; it does not post transaction data to Ethereum for security and is therefore excluded from the L2 leaderboards on this page (also [L2BEAT](https://l2beat.com) categorizes it as "other"). Polygon zkEVM is a ZK rollup and would qualify. ### How many L2s are included in the ranking? 25 chains. The full list (computed on 2026-05-27 UTC from growthepie's master chain catalogue) is: arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. A chain only appears in a given ranking if it has data for that metric — chains without stablecoin coverage simply don't show up. ### Where does this answer come from? Supply comes from growthepie's per-chain stables_mcap timeseries (`/v1/metrics/chains/{chain}/stables_mcap.json`), USD column. Transactions and gas spent come from the blockspace category endpoint (`/v1/blockspace/category_comparison.json`, path `data.token_transfers.subcategories.stablecoin`) — daily uses the latest day of `.daily[chain]`, weekly uses `.aggregated["7d"]`, monthly uses `.aggregated["30d"]`. Variety comes from the stablecoins-by-chain quick-bite table (`/v1/quick-bites/stablecoins/chains/table_{chain}.json`) — the row count is the number of distinct stablecoins on the chain. L2 membership comes from `master.json` (chains where `bucket !== "Layer 1"` and `chain_type` indicates an Ethereum rollup or validium). Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. No editorial overrides. ### Why are weekly and monthly transaction counts called 7d and 30d sums? Because the blockspace endpoint reports stablecoin transactions as rolling-window aggregations rather than calendar weeks or months. "Weekly" on this page = total stablecoin transactions over the most recent 7 days; "monthly" = total over the most recent 30 days. This is different from how the most-used-L2 answer page treats weekly DAA (which uses a calendar-week unique-address count). For supply, weekly and monthly use the per-chain timeseries' native period buckets — those are calendar-aligned. ### Where can I see live L2 stablecoin data? growthepie's [stablecoins-by-chain quick-bite](https://www.growthepie.com/quick-bites/stables-by-chain) lists every stablecoin on every supported chain with live supply timeseries. The [stablecoin market cap dashboard](https://www.growthepie.com/fundamentals/stablecoin-market-cap) shows aggregate supply trends across chains, and the [blockspace overview](https://www.growthepie.com/blockspace/chain-overview) shows category-level activity (including stablecoins) per chain. ### How is "Ethereum L2" defined here? An Ethereum Layer 2 is a chain that derives security from Ethereum by posting transaction data and/or state to Ethereum mainnet. This includes optimistic rollups, ZK rollups, and Validiums. Sidechains (independent validator sets, like Polygon PoS) are excluded. --- ## How many transactions happen on Ethereum L2s? URL: https://www.growthepie.com/answers/ethereum-l2-transaction-count ### Direct answer As of 2026-05-27 UTC, Ethereum L2s collectively process 3960 TPS live, 26.91M transactions per day, 184.20M per week, 788.43M per month, and 35.97B all-time across the 25-chain curated universe (live TPS includes additional L2s tracked by the ecosystem stream). The single largest contributor today is Base Chain (9.85M, 37% of the L2 total). Data: 2026-05-27 UTC. Live leaderboards: growthepie.com/fundamentals/transaction-count. ### Full answer Short answer (data 2026-05-27 UTC): Ethereum L2s collectively process 3960 TPS live, 26.91M transactions per day, 184.20M per week (7-day rolling), 788.43M per month (30-day rolling), and 35.97B all-time across 25 tracked L2s. The single largest contributor today is Base Chain (9.85M, 37%). Updated daily — every figure on this page is recomputed from growthepie's public API. Live TPS uses the latest completed hour; daily uses the most recent completed day; weekly and monthly are rolling sums; all-time covers every day in the ecosystem-wide series since 2021. Live ecosystem TPS Current L2 ecosystem throughput is 3960 TPS (2026-05-27 UTC). Live TPS uses the same real-time data stream as growthepie's Ethereum Ecosystem page: the ecosystem-wide TPS total minus Ethereum L1's own TPS, leaving the L2-only figure. Live TPS spikes during peak hours and dips overnight, so the figure quoted here is a snapshot — not a daily average. Scope note for live TPS only. This single number covers every L2 the ecosystem stream reports — broader than the 25-chain curated universe used for the daily / weekly / monthly / all-time counts below. We do this deliberately for live TPS so the answer matches what users see on the /ethereum-ecosystem page. The historical counts use the curated universe so the per-chain contributors list and the all-time figure are derived from a stable, deduplicated set of chains. Daily, weekly, monthly, all-time As of 2026-05-27 UTC, Ethereum L2s collectively process 3960 TPS live, 26.91M transactions per day, 184.20M per week, 788.43M per month, and 35.97B all-time across the 25-chain curated universe (live TPS includes additional L2s tracked by the ecosystem stream). - Daily (2026-05-27): 26.91M transactions across all L2s. - Weekly (last 7 days ending 2026-05-27): 184.20M transactions. - Monthly (last 30 days ending 2026-05-27): 788.43M transactions. - All-time (cumulative since 2021): 35.97B transactions. Which L2s contribute the most today? Daily transaction count is highly concentrated. The top three contributors on 2026-05-27 UTC, with each chain's share of the ecosystem daily total: 1. Base Chain (9.85M, 37% of L2 total), 2. OP Mainnet (1.68M, 6.2% of L2 total), 3. MegaETH (1.49M, 5.5% of L2 total). The long tail of smaller L2s contributes the remainder. Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue and filter to chains where , , and the chain key is not on the explicit non-L2 list below. 2. For live TPS, pull for the ecosystem-wide total (the same snapshot the /ethereum-ecosystem page uses) and for Ethereum L1's current TPS — the difference is the L2-only figure. The scope here is intentionally broader than the curated universe used elsewhere on this page: the ecosystem stream covers every L2 growthepie tracks live, including some that aren't part of the curated daily/weekly/monthly universe. We use this wider set for live TPS so the number matches what users see on the /ethereum-ecosystem page. If that pair of SSE-snapshot endpoints is unreachable, fall back to summing per-chain from across the curated L2 universe (the fallback uses the narrower set, so it can read slightly lower than the primary path). 3. For daily / weekly / monthly / all-time totals, pull and read — an ecosystem-wide pre-aggregated daily series. Daily = last row; weekly = sum of last 7 rows; monthly = sum of last 30 rows; all-time = sum of every row. 4. For top contributors, read from the same payload, filter to L2s, sort descending, take the top 5. All values shown on this page were generated on 2026-05-27 UTC from growthepie's public API: - Master chain list (with bucket / chaintype classification): - Ecosystem daily transactions and per-chain rankings: - Live ecosystem TPS snapshot (same source as the /ethereum-ecosystem page): (history[0].tps) minus (data.tps) - Fallback live TPS source if the SSE snapshot is unreachable: per-chain hourly TPS, summed across the L2 universe Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Some supporters operate chains that appear in the contributor list above. Counts are computed mechanically from public API data — chains do not pay for inclusion or placement, and supporters do not receive ranking adjustments or preferential treatment. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. Independent L2 transaction data sources you can compare against include L2BEAT (activity, stage classification, risk analysis) and the chains' own block explorers. Methodologies and chain inclusion lists differ between providers — when totals disagree, comparing the underlying chain list (which chains count? are testnets or ARCHIVED chains included?) is usually more informative than the totals themselves. Which chains are included? The list of 25 chains is computed automatically from and refreshed when growthepie adds or removes coverage: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Ethereum mainnet — it is Layer 1, not Layer 2. - Polygon PoS — a sidechain with its own validator set, not a Layer 2. - Aggregate keys ( , ) — used internally to source the ecosystem totals, but not counted as individual chains in the contributor list. ### FAQ ### How many transactions happen on Ethereum L2s? As of 2026-05-27 UTC, Ethereum L2s collectively process 3960 TPS live, 26.91M per day, 184.20M per week, 788.43M per month, and 35.97B all-time across 25 tracked chains. The single largest contributor today is Base Chain (9.85M, 37%). Live leaderboard: growthepie.com/fundamentals/transaction-count. ### What counts as a transaction here? Any confirmed transaction on a tracked Ethereum L2 — including native transfers, ERC-20 transfers, contract calls, and contract deployments. growthepie's transaction count comes directly from chain RPC and indexer data; one transaction = one entry in a block. Aggregations across L2s simply sum these per-chain counts. ### What is the live (real-time) L2 TPS right now? The current L2 ecosystem throughput is **3960 TPS** (data 2026-05-27 UTC). Live TPS uses the same real-time stream that powers growthepie's [Ethereum Ecosystem page](https://www.growthepie.com/ethereum-ecosystem/metrics) — the ecosystem total minus Ethereum L1's own TPS, leaving the L2-only figure quoted here. **Note:** the live TPS figure covers every L2 the ecosystem stream reports — which is broader than the 25-chain curated universe used for the daily / weekly / monthly / all-time counts further down the page. ### How many L2 transactions happen per day? On the latest completed UTC day (2026-05-27), Ethereum L2s processed approximately **26.91M transactions** in total. That figure is the ecosystem-wide daily series exposed by growthepie's landing-page payload. ### How many L2 transactions happen per week? Over the most recent 7 days ending 2026-05-27 UTC, Ethereum L2s processed approximately **184.20M transactions**. Weekly here is a 7-day rolling sum of the daily ecosystem totals — not a calendar week. ### How many L2 transactions happen per month? Over the most recent 30 days ending 2026-05-27 UTC, Ethereum L2s processed approximately **788.43M transactions**. Monthly here is a 30-day rolling sum of the daily ecosystem totals — not a calendar month. ### How many transactions have Ethereum L2s processed all-time? Cumulatively, since growthepie's coverage began in 2021, Ethereum L2s have processed approximately **35.97B transactions** total (data 2026-05-27 UTC). This is the sum of every day in the ecosystem-wide daily transaction series. ### Which L2 contributes the most transactions today? On 2026-05-27 UTC, the top three contributors to total L2 transactions are 1. Base Chain (9.85M, 37% of L2 total), 2. OP Mainnet (1.68M, 6.2% of L2 total), 3. MegaETH (1.49M, 5.5% of L2 total). Shares are the chain's daily transaction count divided by the ecosystem daily total — they sum to roughly 100% when all chains are counted. ### How concentrated is L2 activity across chains? Highly. The top three L2s by daily transaction count account for the bulk of ecosystem volume today. The single-largest contributor on 2026-05-27 UTC is Base Chain (9.85M, 37%). The long tail of smaller L2s contributes the remainder. ### Is Polygon PoS counted as an L2 here? No. Polygon PoS is a sidechain with its own validator set and is excluded from these ecosystem totals (also [L2BEAT](https://l2beat.com) categorizes it as "other"). Polygon zkEVM is a ZK rollup and is included. The same exclusion list as the other L2 answer pages on growthepie applies here. ### Is Ethereum mainnet included in the L2 total? No. These figures cover Ethereum Layer 2s only. Ethereum mainnet (L1) is a separate chain — if you want to compare L1 vs L2, growthepie tracks Ethereum mainnet throughput and transaction count on the [transaction count dashboard](https://www.growthepie.com/fundamentals/transaction-count). ### How many L2s are included in the ecosystem total? It depends on which figure. The **live TPS** number quoted here includes every L2 reported by growthepie's ecosystem stream — the same set the /ethereum-ecosystem page shows — which is broader than the curated universe used for the rest of this page. The **daily / weekly / monthly / all-time** counts and the **top-contributors** breakdown use a curated universe of **25** chains, computed on 2026-05-27 UTC from growthepie's master chain catalogue: arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. A chain only contributes to those counts if it has data for the relevant period. ### Where does this answer come from? Live TPS uses the same source as growthepie's [Ethereum Ecosystem page](https://www.growthepie.com/ethereum-ecosystem/metrics): the most recent ecosystem-wide TPS snapshot from [sse.growthepie.com/api/history](https://sse.growthepie.com/api/history) minus Ethereum L1's current TPS from `sse.growthepie.com/api/chain/ethereum`. The difference is L2-only ecosystem TPS, and **it includes every L2 the ecosystem stream tracks** — a broader set than the curated universe used for the rest of this page. If those SSE-snapshot endpoints are unreachable the helper falls back to summing per-chain `tps` from [growthepie's fees table](https://api.growthepie.com/v1/fees/table.json) across the curated L2 universe (which is the narrower set — fallback can read slightly lower than the primary path). Daily / weekly / monthly / all-time come from growthepie's pre-aggregated ecosystem series at `landing_page.json` `data.all_l2s.metrics.txcount.daily.data` — weekly = sum of last 7 days, monthly = sum of last 30 days, all-time = sum of every day in the series. Top contributors come from the same `landing_page.json` payload's per-chain `ranking.txcount` values. The curated L2 universe (used for everything except live TPS) comes from `master.json` (chains where `bucket !== "Layer 1"` and `chain_type` indicates an Ethereum rollup or validium). Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. No editorial overrides. ### Why is "weekly" a rolling 7-day sum and not a calendar week? Because the ecosystem-wide series is exposed at daily granularity, and a 7-day rolling sum gives a more responsive headline than waiting until a calendar week closes. It also makes the weekly figure update every day, which matches the way live dashboards and AI answers tend to surface data. If you specifically need calendar-week totals, the per-chain transaction-count endpoint (`/v1/metrics/chains/{chain}/txcount.json`) exposes those as period-native weekly aggregates. ### Why does live TPS not match the daily count divided by 86,400? Because they measure different things. Daily transaction count is the total across the whole day — including quieter hours overnight. Live TPS is the latest completed hour's throughput, which is usually higher during peak hours (when most txs happen) and lower overnight. Dividing daily by 86,400 gives the 24-hour AVERAGE TPS, not the live throughput. ### Where can I see live L2 transaction data? growthepie tracks transaction count, throughput (TPS / Mgas/s), and ecosystem totals live at [growthepie.com/fundamentals/transaction-count](https://www.growthepie.com/fundamentals/transaction-count) and [growthepie.com/fundamentals/throughput](https://www.growthepie.com/fundamentals/throughput). The live fees view at [fees.growthepie.com](https://fees.growthepie.com) shows per-chain hourly TPS in a leaderboard. ### How is "Ethereum L2" defined here? An Ethereum Layer 2 is a chain that derives security from Ethereum by posting transaction data and/or state to Ethereum mainnet. This includes optimistic rollups, ZK rollups, and Validiums. Sidechains (independent validator sets, like Polygon PoS) are excluded. --- ## What are the top apps on Ethereum L2s? URL: https://www.growthepie.com/answers/top-apps-ethereum-l2s ### Direct answer Over the last 30 days, the top apps on Ethereum L2s are: **Redstone Finance** by transactions (28.02M), **Circle Internet Financial, LLC.** by active addresses (2.47M, summed across chains), and **Infinitism (ERC-4337) - Account Abstraction** by gas fees paid ($185.1k). Top 10 by transactions: 1. Redstone Finance (28.02M); 2. Infinitism (ERC-4337) - Account Abstraction (15.66M); 3. Circle Internet Financial, LLC. (15.09M); 4. AWP (15.01M); 5. Tether (14.98M); 6. Uniswap (11.94M); 7. KyberSwap (9.53M); 8. Loot Survivor (6.93M); 9. World Markets (5.15M); 10. Lunarbase (3.35M). Data: 2026-05-27 UTC. Live leaderboards: growthepie.com/applications. ### Full answer Short answer (data 2026-05-27 UTC, last 30 days): The top app on Ethereum L2s by transactions is Redstone Finance (28.02M), by active addresses Circle Internet Financial, LLC. (2.47M), and by gas fees paid Infinitism (ERC-4337) - Account Abstraction ($185.1k). Each ranking covers every L2 the app operates on, summed. Updated daily — every leaderboard on this page is recomputed from growthepie's public applications endpoint. The headline window is the last 30 days; rankings within that window use the most recent completed data growthepie has published. How we measure "top app" We rank Ethereum L2 apps by three independent metrics — each tells a different story, so we present all three: - Transaction count — total transactions sent to the app's labelled contracts across L2s. Best proxy for "how busy" an app is. - Active addresses — unique addresses interacting with the app per chain, summed across chains. Best proxy for "how many users". See the cross-chain caveat in the FAQ — a user on multiple L2s is counted per chain. - Gas fees paid — USD-denominated fees the app's users have paid to L2 sequencers over the window. Best proxy for "economic weight" — how much an app contributes to L2 revenue. Top 10 apps by transactions Over the last 30 days (data 2026-05-27 UTC): 1. Redstone Finance (28.02M); 2. Infinitism (ERC-4337) - Account Abstraction (15.66M); 3. Circle Internet Financial, LLC. (15.09M); 4. AWP (15.01M); 5. Tether (14.98M); 6. Uniswap (11.94M); 7. KyberSwap (9.53M); 8. Loot Survivor (6.93M); 9. World Markets (5.15M); 10. Lunarbase (3.35M). Top 10 apps by active addresses Over the last 30 days (data 2026-05-27 UTC): 1. Circle Internet Financial, LLC. (2.47M); 2. Tether (1.62M); 3. Uniswap (734.0k); 4. Focus Tree (237.5k); 5. Relay (187.9k); 6. Mento Protocol (182.4k); 7. LI.FI (133.2k); 8. 0x Protocol (89.8k); 9. OKX (88.7k); 10. Coinbase (81.6k). Active addresses are summed across chains, so apps spanning many L2s rank higher than a strict unique-user count would suggest. Top 10 apps by gas fees paid Over the last 30 days (data 2026-05-27 UTC): 1. Infinitism (ERC-4337) - Account Abstraction ($185.1k); 2. Uniswap ($129.7k); 3. Extended Exchange ($82.5k); 4. Redstone Finance ($51.3k); 5. OKX ($35.3k); 6. GMX ($29.0k); 7. Circle Internet Financial, LLC. ($28.6k); 8. Blazing ($27.9k); 9. AWP ($25.6k); 10. Tether ($24.0k). Gas fees paid is the cumulative USD-denominated fee the app's users have paid to L2 sequencers — a strong "economic weight" signal. Top 5 apps per category (by transactions) Apps in growthepie's labels system are tagged with a main category (DeFi, NFT, gaming, social, etc.). Below is the top 5 in each category by transaction count, with categories ordered by combined transaction volume. Utility 1. Redstone Finance — 28.02M transactions 2. Infinitism (ERC-4337) - Account Abstraction — 15.66M transactions 3. AWP — 15.01M transactions 4. Let's Get HAI — 2.05M transactions 5. drand — 838.4k transactions Token Transfers 1. Circle Internet Financial, LLC. — 15.09M transactions 2. Tether — 14.98M transactions 3. Mento Protocol — 2.51M transactions 4. Fiat24 — 652.8k transactions 5. Canonic — 515.6k transactions Finance 1. Uniswap — 11.94M transactions 2. KyberSwap — 9.53M transactions 3. World Markets — 5.15M transactions 4. Lunarbase — 3.35M transactions 5. AVNU — 2.58M transactions Social 1. Loot Survivor — 6.93M transactions 2. Focus Tree — 2.38M transactions 3. Celo Labs — 2.18M transactions 4. MineBean — 1.61M transactions 5. Offshore Protocol — 494.8k transactions Cross-Chain 1. ultronfoundationdev — 723.9k transactions 2. LayerZero — 554.1k transactions 3. LI.FI — 390.4k transactions 4. Relay — 383.9k transactions 5. Across — 239.1k transactions Collectibles 1. OpenSea — 201.6k transactions 2. Layer3 — 96.0k transactions 3. MegaCorp — 6.7k transactions 4. Yoink! — 1.7k transactions 5. Chonks — 348 transactions Uncategorized 1. NUTS Finance — 31 transactions Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue and filter to chains where , , and the chain key is not on the explicit non-L2 list below. 2. Pull the apps overview ( ) and drop every row whose is not in the L2 universe (this drops Ethereum mainnet + sidechain rows). 3. For each , sum , , and across the remaining rows. 4. Pull project metadata ( ) and attach each project's display name and main category. 5. Sort the aggregated list three times — by txcount, daa, and gasfeesusd — taking the top 10 of each as the headline lists. 6. Group apps by main category, sort each group by transaction count, take the top 5 per category, and order categories by their combined top-5 transaction volume. All values shown on this page were generated on 2026-05-27 UTC from growthepie's public API: - Master chain list (with bucket / chaintype classification): - Per-(app, chain) rollups: (swap for 1d/7d/90d/365d/max) - Project metadata (display name, maincategory): Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Some supporters operate chains that appear in the rankings above; some apps in the rankings have a token relationship with one or more supporters. Rankings are computed mechanically from public API data — apps don't pay for inclusion or ranking, and supporters don't receive ranking adjustments or preferential treatment. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. Independent app-ranking sources you can compare against include DefiLlama (TVL-focused, broader chain coverage), Dune Analytics (community-built dashboards), and the chains' own ecosystem pages. Methodologies differ — DefiLlama ranks by TVL not activity, growthepie ranks by usage. When rankings disagree it's usually because the underlying metric is different, not the data. Which chains are included? The list of 25 chains is computed automatically from and refreshed when growthepie adds or removes coverage: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Ethereum mainnet — it is Layer 1, not Layer 2. Apps that operate on both L1 and L2s have their L1 activity excluded from these rankings. - Polygon PoS — a sidechain with its own validator set, not a Layer 2. - Aggregate keys ( , ) — not individual chains. ### FAQ ### What are the top apps on Ethereum L2s? Over the last 30 days, the top apps on Ethereum L2s are: **Redstone Finance (28.02M)** by transactions; **Circle Internet Financial, LLC. (2.47M)** by active addresses (summed across chains); and **Infinitism (ERC-4337) - Account Abstraction ($185.1k)** by gas fees paid. Top 10 by transactions: 1. Redstone Finance (28.02M); 2. Infinitism (ERC-4337) - Account Abstraction (15.66M); 3. Circle Internet Financial, LLC. (15.09M); 4. AWP (15.01M); 5. Tether (14.98M); 6. Uniswap (11.94M); 7. KyberSwap (9.53M); 8. Loot Survivor (6.93M); 9. World Markets (5.15M); 10. Lunarbase (3.35M). Data: 2026-05-27 UTC. Live leaderboards: [growthepie.com/applications](https://www.growthepie.com/applications). ### How is "top app" defined here? We rank by three independent metrics and present all three so a single number can't mislead you. **Transaction count** is total transactions sent to the app's contracts across L2s. **Active addresses** is the sum of unique addresses interacting with the app per chain (summed across chains — see the cross-chain caveat below). **Gas fees paid** is the USD-denominated fees the app's users have paid to L2 sequencers. A high ranking by gas fees but low ranking by transactions suggests an app with infrequent but expensive transactions (e.g. heavy compute, large swaps). High by transactions and addresses but low by gas fees suggests cheap, high-frequency activity (e.g. memecoin trading on a low-fee L2). ### Which app has the most transactions on Ethereum L2s? Over the last 30 days, the top app by transactions on Ethereum L2s is **Redstone Finance (28.02M)**. Full top 10 by transactions (across all L2s): 1. Redstone Finance (28.02M); 2. Infinitism (ERC-4337) - Account Abstraction (15.66M); 3. Circle Internet Financial, LLC. (15.09M); 4. AWP (15.01M); 5. Tether (14.98M); 6. Uniswap (11.94M); 7. KyberSwap (9.53M); 8. Loot Survivor (6.93M); 9. World Markets (5.15M); 10. Lunarbase (3.35M). ### Which app has the most active addresses on Ethereum L2s? Over the last 30 days, the top app by active addresses on Ethereum L2s is **Circle Internet Financial, LLC. (2.47M)**. Full top 10 by active addresses (summed across chains; see cross-chain caveat below): 1. Circle Internet Financial, LLC. (2.47M); 2. Tether (1.62M); 3. Uniswap (734.0k); 4. Focus Tree (237.5k); 5. Relay (187.9k); 6. Mento Protocol (182.4k); 7. LI.FI (133.2k); 8. 0x Protocol (89.8k); 9. OKX (88.7k); 10. Coinbase (81.6k). ### Which app pays the most gas fees on Ethereum L2s? Over the last 30 days, the top app by gas fees paid on Ethereum L2s is **Infinitism (ERC-4337) - Account Abstraction ($185.1k)**. Full top 10 by gas fees paid (USD, summed across chains): 1. Infinitism (ERC-4337) - Account Abstraction ($185.1k); 2. Uniswap ($129.7k); 3. Extended Exchange ($82.5k); 4. Redstone Finance ($51.3k); 5. OKX ($35.3k); 6. GMX ($29.0k); 7. Circle Internet Financial, LLC. ($28.6k); 8. Blazing ($27.9k); 9. AWP ($25.6k); 10. Tether ($24.0k). Gas fees paid is a useful "economic weight" signal — apps high on this list are paying real money to operate on L2s and contribute meaningfully to chain revenue. ### What are the top apps by category on Ethereum L2s? We rank the top 5 apps in each main category by transaction count. Categories are sorted by combined transaction count (the largest category first). See the per-category section above for the full breakdown — the page renders one ranked list per category. ### Why are active addresses summed across chains? Because the source data exposes daily active addresses per (app, chain), not per app across chains. Summing across chains is the most we can do without an address-overlap dataset. A user who interacts with the same app on multiple L2s in the same window will be counted once per chain — so an app spanning many chains may rank higher by this metric than its true unique-address count would suggest. The transaction count and gas fees rankings are not affected by this caveat. ### What window are these rankings computed over? The last 30 days. We use this window for the headline lists because it's long enough to smooth out single-day spikes and short enough to reflect current activity. growthepie's [applications dashboard](https://www.growthepie.com/applications) lets you switch between 1d / 7d / 30d / 90d / 1y / max windows interactively. ### Is Polygon PoS counted as an L2 here? No. Polygon PoS is a sidechain with its own validator set and is excluded from these app rankings, matching the rest of the answer pages on growthepie. Polygon zkEVM is a ZK rollup and is included. ### Is Ethereum mainnet activity counted? No. Apps that operate on both Ethereum mainnet and L2s will only have their L2 activity counted in these rankings. Mainnet rows in the source data are filtered out before aggregation. ### How many L2s are included? 25 chains. The full list (computed on 2026-05-27 UTC from growthepie's master chain catalogue) is: arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. An app contributes to the rankings if it has activity on at least one chain in that list. ### Where does this answer come from? App metrics come from growthepie's applications overview endpoint (`/v1/apps/app_overview_30d.json`), which exposes per-(project, chain) rollups of transaction count, active addresses, and gas fees paid (USD) over the chosen window. App metadata — display name and main category — comes from growthepie's projects-filtered labels endpoint (`/v1/labels/projects_filtered.json`). L2 membership comes from `master.json` (chains where `bucket !== "Layer 1"` and `chain_type` indicates an Ethereum rollup or validium). Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. No editorial overrides. ### What if an app is missing from the rankings? Apps need to be in growthepie's labelled-contracts dataset to show up. Unlabelled contracts roll up under "Unlabeled" in the blockspace view but don't appear here as named apps. If you operate an app and want it tracked, see growthepie's [labels portal](https://www.growthepie.com/labels) for the contract-submission flow. ### Why is the "gas fees paid" number much smaller than typical fee discussions? Because this number is what the app's USERS paid in total fees to L2 sequencers over the window — not the cost per transaction. A widely-used L2 app can pay millions of dollars in cumulative fees over 30 days while individual transactions cost fractions of a cent. For per-transaction fee costs, see growthepie's [/answers/lowest-fee-ethereum-l2](/answers/lowest-fee-ethereum-l2). ### Where can I see live app data? growthepie's [applications dashboard](https://www.growthepie.com/applications) lists every tracked app on every supported chain with sortable columns for transactions, active addresses, gas fees, and chain breakdowns. Each app has a detail page at `/applications/{owner_project}` with per-chain charts and metadata. ### How is "Ethereum L2" defined here? An Ethereum Layer 2 is a chain that derives security from Ethereum by posting transaction data and/or state to Ethereum mainnet. This includes optimistic rollups, ZK rollups, and Validiums. Sidechains (independent validator sets, like Polygon PoS) are excluded. --- ## Is Ethereum scaling through L2s? URL: https://www.growthepie.com/answers/is-ethereum-scaling-through-l2s ### Direct answer Yes. Data 2026-05-27 UTC: L2s produce 8.20× more transactions than Ethereum mainnet daily (19.40M vs 2.37M), and 20.1× more throughput daily (50.4 Mgas/s vs 2.51 Mgas/s). Weekly and monthly windows show the same direction. Live leaderboards: growthepie.com/fundamentals/throughput. ### Full answer Short answer (data 2026-05-27 UTC): Yes. As of 2026-05-27 UTC, Ethereum L2s collectively process 8.20× more transactions and 20.1× more throughput (gas/s) than Ethereum mainnet alone — every day. The gap holds across weekly and monthly windows too. Updated daily — every ratio on this page is recomputed from growthepie's public per-chain timeseries. Daily uses the latest completed UTC day; weekly and monthly use the most recent completed period. Transactions: L2 ecosystem vs Ethereum mainnet As of 2026-05-27 UTC, Ethereum L2s collectively produce 8.20× more transactions than mainnet daily (19.40M vs 2.37M), 8.80× more weekly (138.07M vs 15.69M), and 7.47× more monthly (544.10M vs 72.83M). - Daily (2026-05-27): 8.20× (L2s 19.40M vs Ethereum 2.37M). - Weekly: 8.80× (L2s 138.07M vs Ethereum 15.69M). - Monthly: 7.47× (L2s 544.10M vs Ethereum 72.83M). Throughput: L2 ecosystem vs Ethereum mainnet As of 2026-05-27 UTC, Ethereum L2s collectively produce 20.1× more throughput than mainnet daily (50.4 Mgas/s vs 2.51 Mgas/s), 19.5× more weekly (49.0 Mgas/s vs 2.52 Mgas/s), and 20.2× more monthly (50.9 Mgas/s vs 2.52 Mgas/s). - Daily (2026-05-27): 20.1× (L2s 50.4 Mgas/s vs Ethereum 2.51 Mgas/s). - Weekly: 19.5× (L2s 49.0 Mgas/s vs Ethereum 2.52 Mgas/s). - Monthly: 20.2× (L2s 50.9 Mgas/s vs Ethereum 2.52 Mgas/s). How to read these ratios The L2 / L1 ratio is the L2 ecosystem total divided by Ethereum mainnet alone. A ratio of "8×" means the L2 ecosystem produces eight times as many transactions (or eight times as much throughput) as Ethereum mainnet does on its own. This is the clearest single answer to "is Ethereum scaling through L2s" — if the ratio is well above 1 and growing, the answer is yes. If it were near 1 or declining, the answer would be no. The numbers above speak for themselves. A few caveats worth keeping in mind: - L2 transaction count is partly a function of cheaper fees. Lower per-transaction cost makes more micro-transactions economic, which inflates raw count. Throughput is more robust to this effect because gas is charged per computational unit. - Throughput is gas per second, not transactions per second. A chain processing 1,000 simple transfers per second has the same TPS as a chain processing 1,000 swaps per second, but the latter has much higher throughput. - Mainnet isn't shrinking, L2 growth is additive. Ethereum mainnet transaction count and throughput are roughly flat over the last year. The L2 ecosystem grew alongside, not at mainnet's expense. Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue to enumerate the curated L2 universe (chains where , , not on the explicit non-L2 list). 2. Fetch Ethereum mainnet values from and — the per-chain endpoint exposes period-native daily / weekly / monthly buckets. 3. Fetch the L2 ecosystem aggregate from . If that endpoint is unreachable, fall back to fetching every L2 in the curated universe and summing per-period values. 4. For each (metric, period) pair, compute the L2 / L1 ratio as . A ratio greater than 1 means L2s produce more of that metric than mainnet over the same window. All values shown on this page were generated on 2026-05-27 UTC: - Master chain list (with bucket / chaintype classification): - Ethereum mainnet transactions: - Ethereum mainnet throughput: - L2 ecosystem aggregate transactions: - L2 ecosystem aggregate throughput: Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Some supporters operate L2 chains, but ratios on this page are computed mechanically from public API data — no supporter receives any preferential treatment. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. Independent sources for the L2-vs-mainnet comparison include L2BEAT (TVL + stage classification + chain inclusion list), Etherscan (mainnet-only metrics), and the chains' own block explorers (per-L2 metrics). Methodologies differ — L2BEAT tracks TVL primarily; growthepie tracks usage (txs, throughput). When ratios disagree it's usually because the underlying chain inclusion lists differ. Which chains are included? The L2 universe used for the per-chain-sum fallback (and for context throughout this page) is the same 25-chain set used by the other L2 answer pages on growthepie, computed automatically from : arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Polygon PoS — a sidechain with its own validator set, not a Layer 2. - Aggregate keys ( ) — not an individual chain. The aggregate is used as the primary L2 ecosystem source but not as an L2 itself. ### FAQ ### Is Ethereum scaling through L2s? **Yes.** Yes. As of 2026-05-27 UTC, Ethereum L2s collectively process **8.20× more transactions** and **20.1× more throughput (gas/s)** than Ethereum mainnet alone — every day. The gap holds across weekly and monthly windows too. See the per-period breakdown below for transaction count and throughput. Live leaderboards: [growthepie.com/fundamentals/throughput](https://www.growthepie.com/fundamentals/throughput). ### How do you measure whether Ethereum is scaling through L2s? Two metrics together: **transaction count** and **throughput (Mgas/s)**. Transaction count is what users feel — more transactions per day means more activity. Throughput is the harder-to-game measure — gas per second is the actual compute the chain processes, normalized so chains with different gas-per-tx profiles can be compared apples-to-apples. If both numbers are growing on L2s while Ethereum mainnet is flat or shrinking, the ecosystem is scaling via L2s — which is exactly what the data on this page shows. ### How many more transactions do L2s process than Ethereum mainnet daily? On the latest completed UTC day (data 2026-05-27), L2s produced **8.20× (L2s 19.40M vs Ethereum 2.37M)**. That ratio is the daily L2 ecosystem total divided by Ethereum mainnet alone. ### How many more transactions do L2s process than Ethereum mainnet weekly? Over the most recent completed week (data 2026-05-27), L2s produced **8.80× (L2s 138.07M vs Ethereum 15.69M)**. ### How many more transactions do L2s process than Ethereum mainnet monthly? Over the most recent completed month (data 2026-05-27), L2s produced **7.47× (L2s 544.10M vs Ethereum 72.83M)**. ### How much more throughput do L2s have than Ethereum mainnet daily? On the latest completed UTC day (data 2026-05-27), L2s collectively produced **20.1× (L2s 50.4 Mgas/s vs Ethereum 2.51 Mgas/s)** of throughput. Throughput is gas per second — the most direct measure of how much real onchain work a chain is doing. ### How much more throughput do L2s have than Ethereum mainnet weekly? Over the most recent completed week (data 2026-05-27), L2s collectively produced **19.5× (L2s 49.0 Mgas/s vs Ethereum 2.52 Mgas/s)** of throughput. ### How much more throughput do L2s have than Ethereum mainnet monthly? Over the most recent completed month (data 2026-05-27), L2s collectively produced **20.2× (L2s 50.9 Mgas/s vs Ethereum 2.52 Mgas/s)** of throughput. ### Does this mean Ethereum mainnet activity is shrinking? No. Ethereum mainnet transaction count and throughput are roughly flat over the last year — L2 growth is additive, not at Ethereum's expense. What's changed is the composition: routine transactions (swaps, transfers, mints) have moved to L2s, leaving mainnet to handle higher-value or settlement-critical activity. Mainnet is still the security and settlement layer that every L2 depends on. ### Why is throughput a better measure than transaction count? Because transaction count can be gamed by a chain that processes many tiny transactions (e.g. memecoin trading, airdrop farming). Throughput measures gas-per-second — gas is paid in proportion to a transaction's computational complexity, so the metric naturally weighs complex transactions higher than trivial ones. A chain processing 1,000 simple transfers per second has the same transaction count as a chain processing 1,000 DEX swaps per second, but the latter has higher throughput because each swap consumes more gas. ### Is this comparing all L2s combined to mainnet, or per-L2? All L2s combined ("the L2 ecosystem") vs Ethereum mainnet alone. We sum across every L2 in growthepie's curated universe to produce the L2 totals; mainnet is its own row. So the ratios you see here are "the L2 ecosystem produces X times more than mainnet" — not "any single L2 produces X times more than mainnet". For per-L2 rankings, see [/answers/most-used-ethereum-l2](/answers/most-used-ethereum-l2). ### Is Polygon PoS counted as an L2 here? No. Polygon PoS is a sidechain with its own validator set and is excluded from the L2 ecosystem totals on this page, matching the rest of the answer pages on growthepie. Polygon zkEVM is a ZK rollup and is included. ### How many L2s are included? 25 chains. The full list (computed on 2026-05-27 UTC from growthepie's master chain catalogue) is: arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. ### Where does this answer come from? Both Ethereum mainnet and L2 ecosystem values come from growthepie's per-chain timeseries endpoints (`/v1/metrics/chains/{chain}/txcount.json` and `/v1/metrics/chains/{chain}/throughput.json`). For Ethereum mainnet the chain key is `ethereum`; for the L2 ecosystem aggregate the helper first tries the `all_l2s` aggregate chain key and falls back to summing every L2 in the curated universe if that endpoint is unreachable. L2 membership comes from `master.json` (chains where `bucket !== "Layer 1"` and `chain_type` indicates an Ethereum rollup or validium). Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. No editorial overrides. ### Why aren't the L2 transaction counts here the same as the cumulative number on /answers/ethereum-l2-transaction-count? Different questions, different aggregations. The cumulative answer page shows the **sum of every day** since 2021 ("all-time"). This page compares L2 vs mainnet for a single period at a time (daily / weekly / monthly), so the L2 numbers here are per-window totals, not cumulative. The daily L2 figure on both pages should match (both use the most recent completed day). ### Where can I see this comparison visually? growthepie's [transaction count dashboard](https://www.growthepie.com/fundamentals/transaction-count) and [throughput dashboard](https://www.growthepie.com/fundamentals/throughput) both let you overlay the `all_l2s` aggregate and `ethereum` series on the same chart for any timespan. The visual divergence between the two lines is the clearest answer to "is Ethereum scaling through L2s". ### How is "Ethereum L2" defined here? An Ethereum Layer 2 is a chain that derives security from Ethereum by posting transaction data and/or state to Ethereum mainnet. This includes optimistic rollups, ZK rollups, and Validiums. Sidechains (independent validator sets, like Polygon PoS) are excluded. --- ## What is the most used Ethereum L2? URL: https://www.growthepie.com/answers/most-used-ethereum-l2 ### Direct answer By throughput, the daily leader is Base Chain (21.5 Mgas/s); weekly leader: Base Chain (20.0 Mgas/s); monthly leader: Base Chain (21.0 Mgas/s). By transaction count, the daily leader is Base Chain (9.85M); weekly leader: Base Chain (67.82M); monthly leader: Base Chain (235.39M). By active addresses (unique per period for weekly/monthly, not summed daily values), the daily leader is Base Chain (466.1k); weekly leader: Base Chain (1.69M); monthly leader: Base Chain (6.15M). Data: 2026-05-27 UTC. Live leaderboards: growthepie.com/fundamentals/throughput. ### Full answer Short answer (data 2026-05-27 UTC): By throughput the leading Ethereum L2 is Base Chain (21.5 Mgas/s). By daily transactions: Base Chain (9.85M). By daily active addresses: Base Chain (466.1k). Weekly and monthly leaders may differ — see the per-metric leaderboards below. Updated daily — every leaderboard and dataset on this page is recomputed from growthepie's public API. Daily values use the latest available day; weekly and monthly values use the most recent completed period (not the in-progress one). How we measure "most used" There are three main metrics we use to measure L2 usage: - Throughput (Mgas/s) — gas processed per second on the chain, like a speedometer for real onchain work. Hard to inflate with cheap spam because every operation costs gas proportional to its complexity. - Transaction count — raw count of transactions per period. Most intuitive but biased toward chains with very low fees and very small transaction sizes (e.g. cheap micro-payments). - Active addresses — count of unique addresses that transacted in the period. Best proxy for users, but a single person can hold many addresses and airdrops can briefly inflate the number. Weekly and monthly values count each address once even if it transacted on multiple days. Live leaderboard: Throughput As of 2026-05-27 UTC, Base Chain leads Ethereum L2 throughput at 21.5 Mgas/s daily, 20.0 weekly, 21.0 monthly. Daily top 3: 1. Base Chain (21.5 Mgas/s), 2. MegaETH (8.05 Mgas/s), 3. OP Mainnet (5.56 Mgas/s). Weekly top 3: 1. Base Chain (20.0 Mgas/s), 2. MegaETH (8.40 Mgas/s), 3. OP Mainnet (5.08 Mgas/s). Monthly top 3: 1. Base Chain (21.0 Mgas/s), 2. OP Mainnet (7.51 Mgas/s), 3. MegaETH (6.96 Mgas/s). Live leaderboard: Transactions As of 2026-05-27 UTC, Base Chain leads Ethereum L2 transaction count at 9.85M daily, 67.82M weekly, 235.39M monthly. Daily top 3: 1. Base Chain (9.85M), 2. OP Mainnet (1.68M), 3. MegaETH (1.49M). Weekly top 3: 1. Base Chain (67.82M), 2. MegaETH (11.15M), 3. OP Mainnet (10.86M). Monthly top 3: 1. Base Chain (235.39M), 2. Arbitrum One (49.85M), 3. OP Mainnet (48.98M). Live leaderboard: Active Addresses As of 2026-05-27 UTC, Base Chain leads Ethereum L2 active addresses at 466.1k daily, 1.69M weekly, 6.15M monthly. Daily top 3: 1. Base Chain (466.1k), 2. Celo (309.0k), 3. Arbitrum One (133.3k). Weekly top 3: 1. Base Chain (1.69M), 2. Celo (661.6k), 3. Arbitrum One (526.4k). Monthly top 3: 1. Base Chain (6.15M), 2. Arbitrum One (3.30M), 3. Celo (1.90M). Methodology and data sources How the answer is derived (transparent methodology): 1. Pull the master chain catalogue and filter to chains where , , and the chain key is not on the explicit non-L2 list below. 2. For each L2 in the universe, pull the per-chain metric endpoint ( ) which exposes the daily / weekly / monthly aggregations natively. Daily uses the latest available data point; weekly and monthly use the most recent completed period. 3. For weekly and monthly active addresses we use the API's period-native aggregate (unique addresses transacting in the window) — not a sum of daily DAAs — so a wallet that transacts on multiple days within a period is only counted once. 4. Sort the chains by raw value for each (metric, period) pair and take the top 3. All rankings on this page pull live from growthepie's public API and refresh daily; the values shown above were generated on 2026-05-27 UTC: - Master chain list (with bucket / chaintype classification): - Per-chain rankings used to compute the leaderboards: (path ) - Throughput time series: - Transaction count time series: - Daily active addresses time series: Data is licensed CC BY-NC 4.0. Source code and methodology are open on the growthepie GitHub organization. Funding disclosure. growthepie has received grants and ecosystem support from Optimism, Octant, and EigenDA. Some supporters operate chains that appear in the rankings above. Rankings are computed mechanically from public API data — chains do not pay for inclusion or placement, and supporters do not receive ranking adjustments or preferential treatment. Full list of supporters and current funding rounds: growthepie.com/donate. Cross-check this answer. Independent L2 data sources you can compare against include L2BEAT (activity, stage classification, risk analysis), and DeFiLlama (TVL across chains). Methodologies and chain inclusion lists differ between providers — when rankings disagree, comparing the underlying definitions is usually more informative than the ranks themselves. Which chains are included? While growthepie tracks the TPS of every L2 in the Ethereum ecosystem, our detailed leaderboards focus on the most widely used and adopted chains. The list of 25 chains is computed automatically from and refreshed when growthepie adds or removes coverage: arbitrum, arbitrumnova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksyncera. What we exclude and why: - Ethereum mainnet — it is Layer 1, not Layer 2. - Polygon PoS — a sidechain with its own validator set, not a Layer 2. - Aggregate keys ( , ) — not individual chains. ### FAQ ### What is the most used Ethereum L2? It depends on the metric and the time horizon. As of 2026-05-27 UTC: by throughput (gas processed per second) the daily leader is Base Chain (21.5 Mgas/s); by raw daily transactions: Base Chain (9.85M); by daily active addresses: Base Chain (466.1k). Weekly and monthly leaders may differ — see the metric-specific FAQs below. All rankings are computed daily from growthepie's data; live leaderboards: growthepie.com/fundamentals/throughput. ### Why does the answer depend on the metric? Different metrics measure different things. Throughput (Mgas/s) is hardest to game and best reflects sustained workload. Transaction count is intuitive but inflates chains with small, cheap transactions. Daily active addresses approximate users but addresses are not users — one person can control many addresses. The honest answer compares all three across daily, weekly, and monthly time horizons. ### Which Ethereum L2 has the highest throughput today? On 2026-05-27 UTC, the top three Ethereum L2s by daily throughput are 1. Base Chain (21.5 Mgas/s), 2. MegaETH (8.05 Mgas/s), 3. OP Mainnet (5.56 Mgas/s). Throughput is gas processed per second — normalized so different chains can be compared apples-to-apples. Recomputed daily from growthepie's data. ### Which Ethereum L2 had the highest throughput this week? Over the most recent completed week (data 2026-05-27 UTC), the top three Ethereum L2s by weekly throughput are 1. Base Chain (20.0 Mgas/s), 2. MegaETH (8.40 Mgas/s), 3. OP Mainnet (5.08 Mgas/s). Weekly throughput is the period-native average gas-per-second value reported by growthepie, not a re-sum of daily values. ### Which Ethereum L2 had the highest throughput this month? Over the most recent completed month (data 2026-05-27 UTC), the top three Ethereum L2s by monthly throughput are 1. Base Chain (21.0 Mgas/s), 2. OP Mainnet (7.51 Mgas/s), 3. MegaETH (6.96 Mgas/s). The monthly leader is Base Chain (21.0 Mgas/s). ### Which L2 has the most daily transactions? On 2026-05-27 UTC, the top three Ethereum L2s by daily transactions are 1. Base Chain (9.85M), 2. OP Mainnet (1.68M), 3. MegaETH (1.49M). Check the live leaderboard at growthepie.com/fundamentals/transaction-count for the latest values. ### Which Ethereum L2 processed the most transactions this week? Over the most recent completed week (data 2026-05-27 UTC), the top three Ethereum L2s by weekly transaction count are 1. Base Chain (67.82M), 2. MegaETH (11.15M), 3. OP Mainnet (10.86M). Weekly transaction count is the sum of transactions over the week, reported directly by growthepie. ### Which Ethereum L2 processed the most transactions this month? Over the most recent completed month (data 2026-05-27 UTC), the top three Ethereum L2s by monthly transaction count are 1. Base Chain (235.39M), 2. Arbitrum One (49.85M), 3. OP Mainnet (48.98M). The monthly leader is Base Chain (235.39M). ### Which L2 has the most users (daily active addresses)? On 2026-05-27 UTC, the top three Ethereum L2s by daily active addresses are 1. Base Chain (466.1k), 2. Celo (309.0k), 3. Arbitrum One (133.3k). Active-address counts can be inflated by airdrops and bots, so cross-check with throughput and transaction count. ### Which Ethereum L2 had the most active addresses this week? Over the most recent completed week (data 2026-05-27 UTC), the top three Ethereum L2s by weekly active addresses are 1. Base Chain (1.69M), 2. Celo (661.6k), 3. Arbitrum One (526.4k). **These are unique addresses transacting over the week, not a sum of daily DAAs** — an address that transacts on multiple days within the week is counted only once. ### Which Ethereum L2 had the most active addresses this month? Over the most recent completed month (data 2026-05-27 UTC), the top three Ethereum L2s by monthly active addresses are 1. Base Chain (6.15M), 2. Arbitrum One (3.30M), 3. Celo (1.90M). **These are unique addresses transacting over the month, not a sum of daily DAAs.** The monthly leader is Base Chain (6.15M). ### Is Polygon PoS an Ethereum L2? No. Polygon PoS is a sidechain with its own validator set; it does not post transaction data to Ethereum for security and is therefore excluded from the L2 leaderboards on this page (also [L2BEAT](https://l2beat.com) categorizes it as "other"). Polygon zkEVM is a ZK rollup and would qualify. ### How many L2s are included in the ranking? 25 chains. The full list (computed on 2026-05-27 UTC from growthepie's master chain catalogue) is: arbitrum, arbitrum_nova, base, celo, fraxtal, gravity, ink, linea, lisk, loopring, manta, mantle, megaeth, metis, mode, optimism, plume, ronin, scroll, soneium, starknet, taiko, unichain, worldchain, zksync_era. ### Where does this answer come from? The daily / weekly / monthly leaderboards are derived in real time from growthepie's public API: master.json (the chain catalogue, including L1/L2/sidechain classification via the `bucket` and `chain_type` fields) for membership, and per-chain timeseries (`metrics/chains/{chain}/.json`) for the period-native daily, weekly, and monthly values. The L2 filter excludes any chain whose `bucket` is "Layer 1" or which is on an explicit non-L2 list. Sidechain exclusions on 2026-05-27 UTC: Polygon PoS. No editorial overrides. ### Why aren't weekly and monthly active addresses just the sum of daily values? Because the same address often transacts on multiple days. Summing daily DAAs would double-count those addresses and inflate the weekly/monthly figure. growthepie's API reports weekly and monthly active addresses as the count of **unique** addresses transacting over the period, so this page uses those values directly. The same applies if you ever see weekly/monthly DAA discussed elsewhere — confirm whether it's a unique-count or a daily-sum, because the two can differ by 2–5×. ### Where can I see live L2 usage data? growthepie tracks every major Ethereum L2 with live throughput, transaction count, daily active addresses, fees, and more. Start at growthepie.com/fundamentals/throughput and use the chain pages (e.g. /chains/base, /chains/arbitrum) for per-chain deep dives. ### How is "Ethereum L2" defined here? An Ethereum Layer 2 is a chain that derives security from Ethereum by posting transaction data and/or state to Ethereum mainnet. This includes optimistic rollups, ZK rollups, and Validiums. Sidechains (independent validator sets, like Polygon PoS) are excluded. --- Generated: 2026-05-27 UTC Index: https://www.growthepie.com/llms.txt