Skip to main content

What Is the OP Stack? Architecture and Superchain Explained

The OP Stack is Optimism's open-source rollup framework powering Base, Mode, and 40+ chains in the Superchain. How it works, who uses it, and what it costs.

Written by Eco

What Is the OP Stack?

The OP Stack is an open-source, modular software framework developed by OP Labs that lets developers deploy Ethereum-equivalent optimistic rollups with standardized components. It defines every layer of a rollup (from transaction sequencing to state root posting on L1) and serves as the shared technical foundation for the Superchain, Optimism's network of interoperable chains.

Before the OP Stack, launching an L2 meant forking Optimism's monolithic node software and maintaining a divergent codebase indefinitely. The OP Stack changed that by separating each function into a discrete module with a published interface. Teams can swap individual modules (the execution engine, the sequencer client, the data availability layer) without forking everything else. Every chain that builds on the standard spec remains composable with every other chain in the network.

The framework is versioned and upgraded through a coordinated protocol governance process. When OP Labs ships a major upgrade, such as the Bedrock rewrite in 2023 or the Fault Proofs activation in 2024, all Superchain members receive the upgrade through the shared contract system rather than through individual chain operator action. This coordination model is one of the primary reasons teams choose the OP Stack over maintaining a fully custom L2 codebase.

As of 2025, the OP Stack powers more than 40 production chains, including Base (Coinbase's L2), Mode, Zora, Redstone, and OP Mainnet itself. Together those chains process several hundred million dollars in daily onchain volume. The codebase is maintained under the ethereum-optimism GitHub org and governed by the Optimism Collective.

The OP Stack is not the only modular rollup framework in production. Arbitrum Orbit and ZK Stack offer competing approaches. The comparison section below covers the five key dimensions across all three. For context on why rollups exist in the first place, see What Is a Rollup? Optimistic vs ZK Rollups.

How Does the OP Stack Architecture Work?

The OP Stack decomposes a rollup into four functional roles. The sequencer collects and orders user transactions. The batcher compresses batches of ordered transactions and posts them to Ethereum L1 as calldata or blobs. The proposer periodically publishes an output root (a cryptographic commitment to the current L2 state) to the L1 contract. The challenger watches for invalid output roots and submits fault proofs within a seven-day window to prevent theft.

Under the hood, the production stack runs three primary services:

op-geth (Execution Engine)

op-geth is a lightly modified fork of go-ethereum that processes EVM transactions and maintains the L2 state database. The modifications are minimal: a custom fee market, a deposit transaction type for bridged assets, and a consensus shim that takes block inputs from the derivation pipeline rather than from peer gossip. Because the diff from upstream geth is small and regularly rebased, OP Stack chains inherit Ethereum's full EVM opcode set and tooling ecosystem with negligible lag.

op-node (Rollup Node)

op-node implements the derivation pipeline: it reads batches from L1, re-derives the canonical L2 block sequence, and feeds those blocks to op-geth via the Engine API. It also runs the P2P layer that lets sequencers gossip unsafe blocks to full nodes before those blocks land onchain. The derivation rule is deterministic: any node that reads the same L1 data arrives at the same L2 state, which is what makes the rollup trustless.

op-batcher and op-proposer

op-batcher submits compressed transaction batches to L1. It uses a channel compression algorithm (currently zlib, with Brotli support available in later upgrades) to minimize data costs. op-proposer posts output roots to the L2OutputOracle contract at regular intervals, typically every one to two hours in production deployments. When EIP-4844 blob transactions are active, op-batcher posts batches as blobs rather than calldata, cutting data costs by roughly 10x compared to pre-Dencun rates.

The system contracts on L1 (OptimismPortal, L1CrossDomainMessenger, L1StandardBridge, and L2OutputOracle) are deployed from a shared factory and receive security upgrades through the Optimism Collective's governance process rather than through individual chain operators. This is a meaningful operational simplification for teams running Superchain members.

To understand how fault proofs protect this system against invalid state roots, see What Are Fault Proofs in Blockchain?

What Is the Superchain?

The Superchain is Optimism's network of OP Stack chains that share a common L1 security anchor (Ethereum), a native cross-chain messaging protocol (the CrossL2Inbox), and a collective economic model in which a fraction of each chain's sequencer revenue flows to the Optimism Collective. A Superchain member is not just a chain that uses the OP Stack. It is a chain that has opted into the shared upgrade path and revenue-sharing agreement.

The economic logic is straightforward. Chains generate sequencer revenue when users pay L2 transaction fees above the cost of posting data to L1. Superchain members commit a share of that net revenue to the Optimism Collective, which funds public goods and protocol R&D. In 2024, Base committed to contributing 15% of its sequencer revenue or 2.5% of its OP token issuance (whichever is larger) to the Collective under the terms published by Coinbase and OP Labs.

On the technical side, Superchain interoperability is being built through two primitives:

  • CrossL2Inbox: A predeployed contract that validates cross-chain message inclusion proofs without waiting for the L1 finality window. In the initial design, messages are "unsafe" until finalized but can be used by applications that accept soft finality.

  • SuperchainTokenBridge: A standard bridge contract that allows native token transfers between Superchain members without wrapping, using the CrossL2Inbox for validity.

The Superchain architecture is documented in the Optimism Interop Specification. L2Beat maintains an independent risk tracker for each Superchain member at l2beat.com.

What Chains Use the OP Stack?

More than 40 production chains run the OP Stack as of 2025. The most widely used are Base, Mode, Zora, Redstone, and OP Mainnet itself. Each chain shares the same core derivation and fault-proof logic while differing in sequencer policy, fee token, and data availability configuration. The ecosystem spans DeFi, NFTs, gaming, enterprise payments, and real-world asset issuance.

Base

Base is the highest-volume OP Stack chain, launched by Coinbase in August 2023. It runs standard OP Stack with no modifications to the execution layer. Base is a Superchain member, processes more daily transactions than OP Mainnet, and is a primary target for USDC issuance via Circle's Cross-Chain Transfer Protocol. Coinbase has committed sequencer revenue to the Optimism Collective under the terms described above.

Mode

Mode is an OP Stack chain launched in January 2024 and optimized for DeFi. Mode introduced a "sequencer fee sharing" model in which applications and users who refer onchain volume receive a rebate from sequencer revenue. This economic layer is built on top of the OP Stack without modifying the protocol itself. Sequencer fee sharing is tracked onchain and settled automatically.

Zora

Zora Network is an OP Stack chain focused on NFT minting and media provenance, launched in June 2023. It processes high volumes of low-value transactions (NFT mints) that benefit disproportionately from the OP Stack's low fixed cost per transaction. Zora is a Superchain member and contributes sequencer revenue to the Collective.

Redstone

Redstone is an OP Stack chain that uses Celestia as an alternative data availability layer rather than Ethereum calldata or blobs. This is possible because the OP Stack's DA module is configurable. Redstone's architecture demonstrates the modular swap OP Labs designed for and targets enterprise and real-world asset use cases where data cost predictability matters more than Ethereum DA guarantees.

OP Mainnet

OP Mainnet is the original Optimism L2 and the reference implementation of the OP Stack. It was the first chain to ship permissionless fault proofs in June 2024, removing the training-wheels multisig that previously controlled dispute resolution. OP Mainnet's onchain activity is tracked in real time at L2Beat.

For a deeper look at Optimism specifically, see What Is Optimism?

OP Stack vs Arbitrum Orbit vs ZK Stack

Three rollup frameworks dominate the L2 deployment landscape in 2025: OP Stack (Optimism), Arbitrum Orbit (Offchain Labs), and ZK Stack (Matter Labs / zkSync). They differ across proof system, EVM compatibility surface, withdrawal latency, data availability options, and ecosystem model. The comparison below covers the five dimensions most relevant to chain operators and application developers selecting a framework.

Dimension

OP Stack

Arbitrum Orbit

ZK Stack

Proof system

Optimistic fault proofs (7-day window); ZK backend in testnet via OP Succinct

Optimistic (Arbitrum BOLD fraud proofs); AnyTrust available for lower-security configs

ZK validity proofs (Boojum STARK); no withdrawal delay after proof verification

EVM compatibility

EVM-equivalent (op-geth fork); identical bytecode, identical gas, direct tooling reuse

EVM-equivalent (Nitro); Stylus adds WASM contracts as an extension

EVM-compatible (zkEVM type 2.5); minor opcode differences, mostly transparent to Solidity devs

Withdrawal latency

7 days (fault proof window); third-party fast bridges available

7 days (fraud proof window) for Rollup mode; AnyTrust is faster with different trust model

Hours to ~24 hours (ZK proof generation plus L1 finality); no fraud window needed

Data availability

Ethereum blobs (default), calldata, Celestia, EigenDA

Ethereum blobs, calldata, AnyTrust DAC (off-chain committee), Celestia

Ethereum blobs, calldata; zkPorter hybrid in development

Network model

Superchain: shared governance, shared upgrade path, sequencer revenue sharing to Collective

Orbit ecosystem: independent chains, no formal revenue sharing, benefits from Arbitrum One liquidity

Hyperchain framework: ZK Stack interop spec in development; no production revenue sharing

Arbitrum Orbit chains do not participate in a shared revenue model the way Superchain members do, but they benefit from Arbitrum One's established DeFi liquidity. ZK Stack chains gain cryptographic finality without a fraud window but carry higher prover infrastructure costs. All three frameworks are open-source. For context on Arbitrum, see What Is Arbitrum?

How Do Developers Deploy an OP Stack Chain?

Deploying an OP Stack chain involves three phases: configuring chain parameters, deploying L1 contracts, and running the sequencer services. The deployment path uses a monorepo-level script that handles contract deployment and genesis generation in a single pass. Most production operators reach a working chain within a day of following the OP Stack chain operator tutorial.

Key configuration decisions before deployment:

  • Chain ID: Must be unique. The Superchain registry tracks registered chain IDs to prevent collisions.

  • Data availability layer: Ethereum blobs (default), Celestia, or EigenDA. Blob mode is cheapest at moderate transaction volumes; AltDA configurations lower costs further with additional trust assumptions.

  • Sequencer address: The EOA or multisig that signs sequencer batches. Decentralized sequencing is an active R&D area; all production deployments today use a single sequencer.

  • Admin and proposer keys: Separate keys for contract upgrades and L2 output proposals. Best practice is a Gnosis Safe multisig for the admin role.

  • Fault proof configuration: Whether to enable the permissionless fault proof system (available since June 2024) or use the legacy output oracle with a trusted proposer.

After genesis, the operator runs op-geth, op-node, op-batcher, and op-proposer as separate processes or containers. The minimum viable sequencer stack requires four services and access to an L1 RPC. Most operators add a second op-node in verifier mode for redundancy.

Gas costs are dominated by L1 data posting. At typical mainnet blob prices in late 2024, a chain posting 100,000 transactions per day to Ethereum incurs L1 data costs in the range of a few hundred dollars per day. This scales with transaction volume and the Ethereum blob base fee. The OP Stack carries no per-chain licensing fee, but Superchain membership requires governance approval from the Optimism Collective.

The ethereum-optimism monorepo is the single source for all services. Chain operators do not need to fork it; the standard pattern is to pin a release tag and supply chain-specific configuration through a TOML or JSON config file. Upgrade migrations are documented in the release notes and typically require a coordinated restart of op-batcher and op-proposer with updated configuration, not a code change.

For context on how L3 chains can be deployed on top of an OP Stack L2, see What Is a Layer 3 (L3) Blockchain?

How Does USDC Work on OP Stack Chains?

USDC operates on OP Stack chains through two mechanisms: bridged USDC and native USDC. Bridged USDC (sometimes labeled USDC.e) is Ethereum USDC locked in the L1StandardBridge with a synthetic representation minted on L2. Native USDC is issued directly by Circle on the L2 and carries native redemption rights. Base and OP Mainnet both have native USDC issued by Circle; smaller OP Stack chains typically start with bridged USDC.

For cross-chain USDC transfers between OP Stack chains, Circle's Cross-Chain Transfer Protocol (CCTP) is the standard bridge. CCTP burns USDC on the source chain, issues an attestation through Circle's off-chain attestation service, and mints native USDC on the destination chain. The round trip typically settles in under five minutes and does not carry the seven-day withdrawal delay of the OP Stack native bridge, because CCTP bypasses the fault proof system entirely.

Verified native USDC contract addresses on major OP Stack chains:

  • Base:0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913

  • OP Mainnet:0x0b2C639c533813f4Aa9D7837CAf62653d097Ff85

Applications building stablecoin payment flows on OP Stack chains should integrate against native USDC contracts and use CCTP for cross-chain transfers rather than the OP Stack's native bridge, which carries the withdrawal delay. For a broader look at stablecoin infrastructure, see 10 Best Stablecoin Automation Platforms 2026.

Why the OP Stack Matters for Onchain Payments

The OP Stack has become the dominant infrastructure layer for onchain stablecoin payments because it combines EVM equivalence, sub-cent transaction fees (at current blob prices), and a growing set of chains that share liquidity and composability through the Superchain architecture. Base alone processes more stablecoin transfer volume than many traditional payment rails handle in emerging markets, and it runs the unmodified OP Stack.

For developers building payment applications, the practical advantages are direct: identical Solidity contracts deploy across all OP Stack chains without modification; native USDC is available on Base and OP Mainnet with Circle CCTP handling cross-chain settlement; and the Superchain's shared messaging layer (when fully live in production) will allow atomic cross-chain transactions, a primitive useful for split payments, multi-chain treasury management, and intent-based settlement.

The same properties that make the OP Stack attractive for consumer applications make it useful for payment infrastructure operators. Low per-transaction cost, EVM equivalence that allows existing Solidity payment contracts to deploy without modification, and a fault-proof security model that inherits Ethereum's L1 security guarantees give OP Stack chains a distinct profile compared to standalone L1 chains or alternative L2 designs. Base's rapid ascent to one of the highest-volume chains for USDC settlement is a direct consequence of these properties.

Eco's routing layer treats OP Stack chains as first-class destinations for stablecoin transfers, with Hyperlane providing cross-chain messaging and CCTP handling USDC bridge flows between supported chains.


Frequently Asked Questions

Is the OP Stack free to use?

The OP Stack codebase is open-source under an MIT license and carries no per-chain licensing fee. Operators pay Ethereum L1 gas to post batches and output roots, plus infrastructure costs for running sequencer services. Superchain membership involves a sequencer revenue-sharing commitment to the Optimism Collective but is not required to use the software itself.

What is the difference between the OP Stack and Optimism?

Optimism refers to OP Mainnet, the original Optimism L2 chain. The OP Stack is the open-source framework that powers OP Mainnet and dozens of other chains. OP Labs, the development company behind both, extracted the OP Stack from the OP Mainnet codebase and generalized it so any team can deploy their own chain using the same proven software.

Does the OP Stack support ZK proofs?

The current production OP Stack uses optimistic fault proofs with a seven-day challenge window. OP Labs and the OP Succinct project are developing a ZK proof backend using SP1, a RISC-V ZK virtual machine from Succinct Labs, that would replace the fraud proof window with near-instant validity proofs. As of early 2025, this runs on testnets but has not shipped to OP Mainnet or Base in production.

How many chains are in the Superchain?

The Superchain registry lists over 40 chains that have adopted the OP Stack and opted into shared upgrade governance as of 2025. This includes Base, Mode, Zora, Redstone, Fraxtal, Metal, and others. The count continues to grow as the Superchain's native interoperability features approach production readiness and chain deployment costs remain low.

What causes the seven-day withdrawal delay on OP Stack chains?

OP Stack withdrawals from L2 to Ethereum L1 require a seven-day waiting period because that is the fault proof challenge window. It gives challengers time to submit a fraud proof if the proposed output root is invalid. Once the window passes without a successful challenge, the withdrawal is finalized and funds can be claimed on L1. Third-party liquidity providers offer instant bridge exits by fronting funds in exchange for a small fee.


Sources: Optimism Protocol Specs; ethereum-optimism/optimism on GitHub; L2Beat risk assessments; Circle CCTP developer documentation. Chain contract addresses sourced from official deployment registries. Cost and volume figures cited as order-of-magnitude estimates consistent with publicly available L2Beat and Dune data as of Q1 2025.

Did this answer your question?