Optimism Bridge vs Across: Capital Efficiency

Native, trust-minimized transfers: Uses the official Optimism Bedrock message-passing layer. This matters for large-value institutional transfers where security is paramount, as funds are never custodied by a third party.

Higher latency for lower cost: Finality is slow (7-day challenge period), but transaction fees are minimal (just L1 gas). This matters for non-time-sensitive treasury operations or bulk fund migrations where cost predictability is key.

Near-instant finality: Uses a bonded relay network and UMA's optimistic oracle to settle in minutes, not days. This matters for arbitrage, trading, and user-facing apps where speed is critical for capital velocity.

Direct canonical bridge: Uses Optimism's native messaging layer for L1<>L2 transfers. This means zero third-party liquidity risk and predictable, low fees paid in ETH/OP. Ideal for protocol treasuries moving large sums or users prioritizing security over instant finality.

Deep ecosystem alignment: As the official bridge, it's natively integrated with core Optimism stack tools like the Superchain Faucet and OP Stack governance. This ensures first-party support and is the required path for sequencer and protocol deployments on the network.

Challenge period delay: The 7-day withdrawal window for standard exits is a significant capital efficiency penalty. While fast bridges exist (like the Canonical Transaction Chain fast path), they add complexity. Not suitable for active trading or arbitrage requiring sub-hour settlement.

Instant liquidity via relayers: Uses a network of bonded relayers and a single-sided liquidity pool model to provide funds in minutes on the destination chain. This eliminates the 7-day wait, making it optimal for arbitrage, collateral repositioning, and time-sensitive transactions.

Dynamic fee model: Fees are based on a real-time capital opportunity cost calculated by the Across UMA oracle, not just gas. For high-volume periods on the destination chain, this can be cheaper than paying for L1 gas directly. Best for users sensitive to total cost, not just speed.

Relayer and liquidity risk: Users trust the economic security of relayers and the sufficiency of the bridge pool TVL (e.g., on Ethereum). While bonds and slashing provide security, it introduces a third-party dependency not present in canonical bridges. Large transfers may face liquidity constraints.

Zero slippage for canonical assets: Transfers of ETH and standard ERC-20s via the official bridge have no price impact, as assets are minted/burned 1:1 on L2. This is critical for large, protocol-to-protocol transfers of native assets like OP, SNX, or wETH.

High latency and capital lockup: Uses a 7-day challenge period for withdrawals to Ethereum mainnet. This ties up capital and is unsuitable for arbitrage, fast portfolio rebalancing, or any time-sensitive operation. The speed is a function of Optimistic Rollup security, not efficiency.

Near-instant finality with bonded liquidity: Uses a model of relayers who front capital, providing users instant receipt of funds on the destination chain. The ~2-5 minute wait is for economic finality, not a security delay. Ideal for arbitrage and reacting to market movements.

Slippage and liquidity fragmentation: Speed requires deep, fragmented liquidity pools on each chain. For large transfers or exotic assets, slippage can be significant. You are trading against a pool, not minting. Performance depends entirely on third-party liquidity providers (LPs).