Major network upgrade featuring Block-level Access Lists and ePBS. Named after the combination of "Amsterdam" (execution layer upgrade, named after the previous Devconnect location) and "Gloas" (consensus layer upgrade, named after a star).
Can subscribe to a standard ETH transfer log instead of tracing; requires small updates to indexers and business logic.
Make every ETH transfer—including from contracts and SELFDESTRUCT—emit an event log. Apps and explorers can track ETH movements uniformly, similar to ERC-20 transfers, improving deposit detection and reducing ad-hoc tracing.
No direct impact. It enables applications to leverage previous builder information, something that is not currently available. It enables trustless preconfirmation schemes.
Proposes the decoupling of the consensus block from the execution payload, both in broadcast and validation. This feature enables L1 scaling by significantly changing the time required to both broadcast and executing the payload together with all the blob data, from the current ~2 seconds to aproximately ~9 seconds. It allows for a maximum portion of the slot to be spent in propagation large data.
Contracts generally unchanged. Only strategies exploiting large refunds lose ability to help pack extra computation into blocks via accounting quirks.
Closes a loophole where gas refunds let blocks pack extra work. Refunds still reduce user costs, but don't shrink the block's counted gas. Aligns block size with real EVM work and reduces worst-case block-size variance.
Can optimize applications for block-level access patterns, significantly reduce gas costs for state-heavy applications, and enable new design patterns.
This introduces access lists at the block level rather than individual transactions, dramatically reducing gas costs for applications that access similar state and enabling new optimization patterns.
Can deploy contracts up to 32KiB and initcode up to 64KiB, reducing pressure to split logic.
Raises limits so developers can deploy slightly larger contracts: up to 32KiB deployed code and 64KiB initcode after a fork. Existing contracts remain valid.
Changes how fee is managed based on accessed state, lowers base cost for every kind of transaction.
Changes base tx cost for more efficiency in cold/hot state management and improving ETH monetary functions. Lowering basic ETH tranfers to 6,000 gas and charging extra 25,000 for new account creation. Alternative to increasing block gas limit.
Making sure devs cannot deploy contract on incorrect address with proper error handling.
Existing conditions to prevent deploying contract at address with existing data are insufficient, this ensures it cannot happen even in edge case scenarios. There are at least 28 contracts that were deployed with this issue.
This EIP stabilizes Merkle tree indices for all consensus data, preventing coincidental verifier breakage when fields are added, removed, or list capacities change during Ethereum upgrades.
No direct impact. Internal networking change only.
When Ethereum processes more transactions per block, the receipts can get too large to send in one network message. This lets nodes request receipts in smaller chunks, preventing sync failures as the network scales.
Application behavior during recovery from mass slashings becomes more predictable, with fewer missed blocks and smoother chain progress. No application changes required.
After a slashing event, punished validators can still be scheduled to propose blocks, causing missed slots and slow recovery. This change simply skips slashed validators so healthy ones keep proposing and the chain keeps running.
Application logic and smart contracts remain unchanged; only underlying consensus churn limits adjust, so most applications require no code modifications.
Raises limits on how many validators can activate, exit, or consolidate each epoch and separates consolidation churn, improving validator consolidation speed and easing deposit and exit queues while preserving a long weak-subjectivity period and existing security assumptions.
Application logic unchanged; contracts or rollup sequencers relying on blob data get steadier L1 availability during busy network-wide periods.
Ethereum nodes fully download only some blob transactions and sample the exact blob pieces the consensus layer cares about, cutting Execution Layer bandwidth roughly fourfold while keeping blob transactions moving.
No protocol changes for smart contract developers; only economic behaviour of validator exits and staking derivatives liquidity may subtly improve.
Changes how validator exits are queued by letting them share capacity with the consolidation queue. Removes an advantage for very large validators and lets more stakers exit or rebalance quickly without weakening security.
No direct impact. Internal networking change only.
Allows Ethereum nodes to share block access lists over the network. These lists record which accounts and storage slots a block touched, enabling faster syncing and parallel transaction processing.
Optionally watch a canonical event for ETH transfers to improve UX; otherwise existing send/receive flows remain unchanged.
Make every ETH transfer—including from contracts and SELFDESTRUCT—emit an event log. Apps and explorers can track ETH movements uniformly, similar to ERC-20 transfers, improving deposit detection and reducing ad-hoc tracing.
No direct impact. It enables wallets to send encrypted txs to anonymous builders.
Proposes the decoupling of the consensus block from the execution payload, both in broadcast and validation. This feature enables L1 scaling by significantly changing the time required to both broadcast and executing the payload together with all the blob data, from the current ~2 seconds to aproximately ~9 seconds. It allows for a maximum portion of the slot to be spent in propagation large data.
No changes required. Gas estimation, fee math, and refund display remain the same at the transaction level for users today.
Closes a loophole where gas refunds let blocks pack extra work. Refunds still reduce user costs, but don't shrink the block's counted gas. Aligns block size with real EVM work and reduces worst-case block-size variance.
Indirect benefits through lower gas costs for applications, but minimal direct impact on wallet development.
This introduces access lists at the block level rather than individual transactions, dramatically reducing gas costs for applications that access similar state and enabling new optimization patterns.
No new transaction types; deployments may include larger bytecode. Any size validations should reflect new limits.
Raises limits so developers can deploy slightly larger contracts: up to 32KiB deployed code and 64KiB initcode after a fork. Existing contracts remain valid.
The fee cost estimation needs to be updated with new changes and complexity.
Changes base tx cost for more efficiency in cold/hot state management and improving ETH monetary functions. Lowering basic ETH tranfers to 6,000 gas and charging extra 25,000 for new account creation. Alternative to increasing block gas limit.
Existing conditions to prevent deploying contract at address with existing data are insufficient, this ensures it cannot happen even in edge case scenarios. There are at least 28 contracts that were deployed with this issue.
This EIP stabilizes Merkle tree indices for all consensus data, preventing coincidental verifier breakage when fields are added, removed, or list capacities change during Ethereum upgrades.
No direct impact. Internal networking change only.
When Ethereum processes more transactions per block, the receipts can get too large to send in one network message. This lets nodes request receipts in smaller chunks, preventing sync failures as the network scales.
Wallets get more consistent head updates after slashings, reducing confusing gaps in block production. No direct wallet code changes are expected.
After a slashing event, punished validators can still be scheduled to propose blocks, causing missed slots and slow recovery. This change simply skips slashed validators so healthy ones keep proposing and the chain keeps running.
Wallet protocols stay the same; only beacon-chain churn parameters change, so implementations rarely need updates beyond optional queue displays.
Raises limits on how many validators can activate, exit, or consolidate each epoch and separates consolidation churn, improving validator consolidation speed and easing deposit and exit queues while preserving a long weak-subjectivity period and existing security assumptions.
Wallets submitting type-3 blob transactions gain more reliable propagation; minimal adjustments, mainly ensuring compatibility with updated EL peers locally.
Ethereum nodes fully download only some blob transactions and sample the exact blob pieces the consensus layer cares about, cutting Execution Layer bandwidth roughly fourfold while keeping blob transactions moving.
Minimal impact. Wallets showing validator exits or withdrawals may adjust messaging on exit timing in congestion; no protocol changes required.
Changes how validator exits are queued by letting them share capacity with the consolidation queue. Removes an advantage for very large validators and lets more stakers exit or rebalance quickly without weakening security.
No direct impact. Internal networking change only.
Allows Ethereum nodes to share block access lists over the network. These lists record which accounts and storage slots a block touched, enabling faster syncing and parallel transaction processing.
Deposits and withdrawals are detected more reliably across apps and exchanges; no change to how users send or receive ETH.
Make every ETH transfer—including from contracts and SELFDESTRUCT—emit an event log. Apps and explorers can track ETH movements uniformly, similar to ERC-20 transfers, improving deposit detection and reducing ad-hoc tracing.
Second order effect of lower prices and higher tx throughput.
Proposes the decoupling of the consensus block from the execution payload, both in broadcast and validation. This feature enables L1 scaling by significantly changing the time required to both broadcast and executing the payload together with all the blob data, from the current ~2 seconds to aproximately ~9 seconds. It allows for a maximum portion of the slot to be spent in propagation large data.
No substantial direct impact; transactions execute as before. Refunds still discount user costs but no longer shrink counted block gas.
Closes a loophole where gas refunds let blocks pack extra work. Refunds still reduce user costs, but don't shrink the block's counted gas. Aligns block size with real EVM work and reduces worst-case block-size variance.
Lower gas costs for complex applications, especially DeFi protocols and applications that access similar state across multiple transactions.
This introduces access lists at the block level rather than individual transactions, dramatically reducing gas costs for applications that access similar state and enabling new optimization patterns.
No required action; may benefit from dapps deploying larger contracts after the network upgrade.
Raises limits so developers can deploy slightly larger contracts: up to 32KiB deployed code and 64KiB initcode after a fork. Existing contracts remain valid.
Lower fees means it's easier to use ETH for payments.
Changes base tx cost for more efficiency in cold/hot state management and improving ETH monetary functions. Lowering basic ETH tranfers to 6,000 gas and charging extra 25,000 for new account creation. Alternative to increasing block gas limit.
Existing conditions to prevent deploying contract at address with existing data are insufficient, this ensures it cannot happen even in edge case scenarios. There are at least 28 contracts that were deployed with this issue.
This EIP stabilizes Merkle tree indices for all consensus data, preventing coincidental verifier breakage when fields are added, removed, or list capacities change during Ethereum upgrades.
No direct impact. Enables future gas limit increases without sync issues.
When Ethereum processes more transactions per block, the receipts can get too large to send in one network message. This lets nodes request receipts in smaller chunks, preventing sync failures as the network scales.
Users should experience fewer stalled slots and better network responsiveness after large validator slashings, because only non-slashed validators will propose blocks.
After a slashing event, punished validators can still be scheduled to propose blocks, causing missed slots and slow recovery. This change simply skips slashed validators so healthy ones keep proposing and the chain keeps running.
No direct user-facing changes; effects are limited to validator churn parameters, with only indirect benefits through future finality improvements.
Raises limits on how many validators can activate, exit, or consolidate each epoch and separates consolidation churn, improving validator consolidation speed and easing deposit and exit queues while preserving a long weak-subjectivity period and existing security assumptions.
No user-facing change; blob transactions stay reliable as BPO forks raise throughput, reducing failures from Execution Layer bandwidth pressure.
Ethereum nodes fully download only some blob transactions and sample the exact blob pieces the consensus layer cares about, cutting Execution Layer bandwidth roughly fourfold while keeping blob transactions moving.
Indirect impact: staking withdrawals and liquid staking tokens may become more responsive during heavy demand, but user-facing interfaces remain unchanged.
Changes how validator exits are queued by letting them share capacity with the consolidation queue. Removes an advantage for very large validators and lets more stakers exit or rebalance quickly without weakening security.
No direct impact. Enables faster node sync and block processing.
Allows Ethereum nodes to share block access lists over the network. These lists record which accounts and storage slots a block touched, enabling faster syncing and parallel transaction processing.
Rollups mirroring L1 EVM semantics may add the same log for consistency; others have no direct requirements.
Make every ETH transfer—including from contracts and SELFDESTRUCT—emit an event log. Apps and explorers can track ETH movements uniformly, similar to ERC-20 transfers, improving deposit detection and reducing ad-hoc tracing.
Second order effect of higher blob count being possible.
Proposes the decoupling of the consensus block from the execution payload, both in broadcast and validation. This feature enables L1 scaling by significantly changing the time required to both broadcast and executing the payload together with all the blob data, from the current ~2 seconds to aproximately ~9 seconds. It allows for a maximum portion of the slot to be spent in propagation large data.
Minimal direct impact. Sequencers mirroring L1 semantics may adopt same accounting; otherwise, no changes to rollup fee or capacity models.
Closes a loophole where gas refunds let blocks pack extra work. Refunds still reduce user costs, but don't shrink the block's counted gas. Aligns block size with real EVM work and reduces worst-case block-size variance.
More efficient state access in Layer 2 settlement transactions and optimized bridge operations.
This introduces access lists at the block level rather than individual transactions, dramatically reducing gas costs for applications that access similar state and enabling new optimization patterns.
Systems mirroring Ethereum’s limits may need updates to contract-size checks to remain compatible with L1 rules.
Raises limits so developers can deploy slightly larger contracts: up to 32KiB deployed code and 64KiB initcode after a fork. Existing contracts remain valid.
L2s aiming to be EVM compatible need to implement this change, ETH will can better for payments even on L2 chains.
Changes base tx cost for more efficiency in cold/hot state management and improving ETH monetary functions. Lowering basic ETH tranfers to 6,000 gas and charging extra 25,000 for new account creation. Alternative to increasing block gas limit.
L2s can benefit from the improvement following the change in their implementations.
Existing conditions to prevent deploying contract at address with existing data are insufficient, this ensures it cannot happen even in edge case scenarios. There are at least 28 contracts that were deployed with this issue.
This EIP stabilizes Merkle tree indices for all consensus data, preventing coincidental verifier breakage when fields are added, removed, or list capacities change during Ethereum upgrades.
No direct impact. May enable higher L1 throughput they can leverage.
When Ethereum processes more transactions per block, the receipts can get too large to send in one network message. This lets nodes request receipts in smaller chunks, preventing sync failures as the network scales.
L2 sequencers relying on L1 data and finality benefit from steadier L1 block production after slashings, improving uptime. No L2 code impact.
After a slashing event, punished validators can still be scheduled to propose blocks, causing missed slots and slow recovery. This change simply skips slashed validators so healthy ones keep proposing and the chain keeps running.
Layer 2 protocols see minimal direct impact; beacon-chain validator churn parameters change, but L2 smart contract logic remains unchanged.
Raises limits on how many validators can activate, exit, or consolidate each epoch and separates consolidation churn, improving validator consolidation speed and easing deposit and exit queues while preserving a long weak-subjectivity period and existing security assumptions.
Rollups posting blobs to L1 face lower risk of EL blobpool saturation, keeping DA publication within expected windows.
Ethereum nodes fully download only some blob transactions and sample the exact blob pieces the consensus layer cares about, cutting Execution Layer bandwidth roughly fourfold while keeping blob transactions moving.
Layer 2 protocols have no direct integration work; systems depending on staking liquidity may benefit from faster exits and rebalancing.
Changes how validator exits are queued by letting them share capacity with the consolidation queue. Removes an advantage for very large validators and lets more stakers exit or rebalance quickly without weakening security.
No direct impact. May benefit from faster L1 finality confirmation.
Allows Ethereum nodes to share block access lists over the network. These lists record which accounts and storage slots a block touched, enabling faster syncing and parallel transaction processing.
More log entries per block; execution node receipts grow slightly, increasing storage and bandwidth requirements for some RPC setups.
Make every ETH transfer—including from contracts and SELFDESTRUCT—emit an event log. Apps and explorers can track ETH movements uniformly, similar to ERC-20 transfers, improving deposit detection and reducing ad-hoc tracing.
Major changes and updates are needed for trustless monitoring in staking pools. Staking UX is improved by a refined builder picking and monitoring.
Proposes the decoupling of the consensus block from the execution payload, both in broadcast and validation. This feature enables L1 scaling by significantly changing the time required to both broadcast and executing the payload together with all the blob data, from the current ~2 seconds to aproximately ~9 seconds. It allows for a maximum portion of the slot to be spent in propagation large data.
Lower variance in execution work per block improves predictability and reduces pathological worst-case loads; efficiency gains from smoother resource usage.
Closes a loophole where gas refunds let blocks pack extra work. Refunds still reduce user costs, but don't shrink the block's counted gas. Aligns block size with real EVM work and reduces worst-case block-size variance.
More efficient block processing due to optimized state access patterns, reducing computational overhead.
This introduces access lists at the block level rather than individual transactions, dramatically reducing gas costs for applications that access similar state and enabling new optimization patterns.
Operators need upgraded clients at fork; larger contracts can slightly increase storage and processing per deployment.
Raises limits so developers can deploy slightly larger contracts: up to 32KiB deployed code and 64KiB initcode after a fork. Existing contracts remain valid.
No direct impact, state growth should stay constrained.
Changes base tx cost for more efficiency in cold/hot state management and improving ETH monetary functions. Lowering basic ETH tranfers to 6,000 gas and charging extra 25,000 for new account creation. Alternative to increasing block gas limit.
Existing conditions to prevent deploying contract at address with existing data are insufficient, this ensures it cannot happen even in edge case scenarios. There are at least 28 contracts that were deployed with this issue.
Fewer required upgrades of staking pool delegate contracts.
This EIP stabilizes Merkle tree indices for all consensus data, preventing coincidental verifier breakage when fields are added, removed, or list capacities change during Ethereum upgrades.
Required upgrade to maintain sync capability as gas limit increases.
When Ethereum processes more transactions per block, the receipts can get too large to send in one network message. This lets nodes request receipts in smaller chunks, preventing sync failures as the network scales.
Once slashed, a validator immediately loses proposer eligibility. Node operators must account for skipped proposing duties but otherwise run unchanged software.
After a slashing event, punished validators can still be scheduled to propose blocks, causing missed slots and slow recovery. This change simply skips slashed validators so healthy ones keep proposing and the chain keeps running.
Stakers and node operators gain higher churn throughput, shortening queues and reducing consolidation costs while maintaining a long weak-subjectivity period.
Raises limits on how many validators can activate, exit, or consolidate each epoch and separates consolidation churn, improving validator consolidation speed and easing deposit and exit queues while preserving a long weak-subjectivity period and existing security assumptions.
Validator operators must run EL clients supporting sparse blobpool to match CL custody; operational workflows stay largely intact.
Ethereum nodes fully download only some blob transactions and sample the exact blob pieces the consensus layer cares about, cutting Execution Layer bandwidth roughly fourfold while keeping blob transactions moving.
Validators and staking providers gain fairer access to extra churn, enabling predictable exit times and increased capacity during rush periods.
Changes how validator exits are queued by letting them share capacity with the consolidation queue. Removes an advantage for very large validators and lets more stakers exit or rebalance quickly without weakening security.
Enables parallel execution optimizations. Storage considerations for BAL retention.
Allows Ethereum nodes to share block access lists over the network. These lists record which accounts and storage slots a block touched, enabling faster syncing and parallel transaction processing.
Explorers and data indexers must parse and surface the new log; moderate engineering to integrate, test, and backfill if desired.
Make every ETH transfer—including from contracts and SELFDESTRUCT—emit an event log. Apps and explorers can track ETH movements uniformly, similar to ERC-20 transfers, improving deposit detection and reducing ad-hoc tracing.
Explorers and block monitors need updates to handle the separation of the beacon block from the payload.
Proposes the decoupling of the consensus block from the execution payload, both in broadcast and validation. This feature enables L1 scaling by significantly changing the time required to both broadcast and executing the payload together with all the blob data, from the current ~2 seconds to aproximately ~9 seconds. It allows for a maximum portion of the slot to be spent in propagation large data.
Block builders, simulators, and analyzers should enforce and reflect gross block-gas accounting; update tests and dashboards to avoid refund-based variance.
Closes a loophole where gas refunds let blocks pack extra work. Refunds still reduce user costs, but don't shrink the block's counted gas. Aligns block size with real EVM work and reduces worst-case block-size variance.
Gas estimation tools, transaction simulation, and optimization analysis need updates to handle block-level access list patterns.
This introduces access lists at the block level rather than individual transactions, dramatically reducing gas costs for applications that access similar state and enabling new optimization patterns.
Explorers, indexers, verifiers, and RPC tooling should handle larger contract bytecode and initcode in deployment flows.
Raises limits so developers can deploy slightly larger contracts: up to 32KiB deployed code and 64KiB initcode after a fork. Existing contracts remain valid.
Transaction analysis needs to distinguish new gas accounting.
Changes base tx cost for more efficiency in cold/hot state management and improving ETH monetary functions. Lowering basic ETH tranfers to 6,000 gas and charging extra 25,000 for new account creation. Alternative to increasing block gas limit.
Contract development tooling needs to update and handle errors correctly
Existing conditions to prevent deploying contract at address with existing data are insufficient, this ensures it cannot happen even in edge case scenarios. There are at least 28 contracts that were deployed with this issue.
Less maintenance for consensus light client components.
This EIP stabilizes Merkle tree indices for all consensus data, preventing coincidental verifier breakage when fields are added, removed, or list capacities change during Ethereum upgrades.
Node operators benefit from more reliable sync at higher gas limits.
When Ethereum processes more transactions per block, the receipts can get too large to send in one network message. This lets nodes request receipts in smaller chunks, preventing sync failures as the network scales.
Indexers, explorers, and relays see steadier block flow after slashings, easing incident handling. No protocol-facing changes required for these services.
After a slashing event, punished validators can still be scheduled to propose blocks, causing missed slots and slow recovery. This change simply skips slashed validators so healthy ones keep proposing and the chain keeps running.
Validator dashboards, explorers, and forecasting tools must update churn constants and formulas to correctly estimate activation, exit, and consolidation queues.
Raises limits on how many validators can activate, exit, or consolidate each epoch and separates consolidation churn, improving validator consolidation speed and easing deposit and exit queues while preserving a long weak-subjectivity period and existing security assumptions.
Block explorers, relays and monitors should surface partial-blob availability and new eth/71 fields to reflect node sampling state.
Ethereum nodes fully download only some blob transactions and sample the exact blob pieces the consensus layer cares about, cutting Execution Layer bandwidth roughly fourfold while keeping blob transactions moving.
Monitoring and analytics tools tracking exit queue lengths must update models to account for churn reallocation via the consolidation queue.
Changes how validator exits are queued by letting them share capacity with the consolidation queue. Removes an advantage for very large validators and lets more stakers exit or rebalance quickly without weakening security.
Node operators may see faster sync times when BALs are available from peers.
Allows Ethereum nodes to share block access lists over the network. These lists record which accounts and storage slots a block touched, enabling faster syncing and parallel transaction processing.