How to Bridge Crypto: 2026 Complete Guide

How do you bridge crypto? Bridging is the process of moving an asset from one blockchain to another. In 2026 there are five main bridge types: native L2 bridges (Arbitrum Bridge, Optimism Gateway, Base Bridge, official, audited, slow withdrawal), liquidity-pool bridges (Across, Stargate, fast, fee-paid), messaging bridges (LayerZero, Wormhole, Axelar, programmable, application-flexible), centralized exchange bridges (deposit on one chain, withdraw on another, custodial but reliable), and native-asset protocols (Circle CCTP for USDC, burn-and-mint with no wrapped token). The honest 2026 framing: bridges remain the dominant DeFi attack vector. April 2026's KelpDAO $292 million LayerZero bridge exploit was the largest DeFi hack of 2026; Peckshield tracked 8 bridge exploits through mid-May with $328.6 million drained cumulatively. Use only audited, mature bridges for material amounts; prefer native-asset routes like Circle CCTP over wrapped-token routes when available.

This guide answers what a user actually needs about how to bridge crypto in 2026: how bridging works mechanically, the five bridge types and how they differ, step-by-step routes for the most common transfers (ETH mainnet to L2, USDC cross-chain via CCTP), the best 2026 bridges by category, crypto bridge fees by route, the wrapped-Bitcoin model, tax treatment, and an honest risk inventory including the 2026 exploit record. Every figure is sourced to a primary citation in the footer.

What is a crypto bridge in 2026?

The short answer to how to bridge crypto: pick a bridge type appropriate to your asset and route, connect a wallet, deposit on the source chain, and receive the asset on the destination chain. A crypto bridge moves value between blockchains that cannot natively communicate. Bitcoin and Ethereum are separate networks with separate ledgers; an asset on Bitcoin is not directly transferable to an Ethereum address. Bridges solve this by locking the asset on the source chain and issuing an equivalent on the destination chain, or by burning the asset on the source and minting on the destination, or by routing through a liquidity pool that absorbs the asset on one side and releases an equivalent on the other.

The 2026 bridge landscape carries approximately $20-25 billion in TVL across all categories, tracked live by DefiLlama's bridge category. The dominant routes serve Ethereum mainnet to L2 transfers (Arbitrum, Base, Optimism), L1-to-L1 transfers (Ethereum to Solana, Ethereum to Cosmos), and stablecoin cross-chain movement (Circle CCTP for USDC across 20+ chains). For broader L2 context including how rollups inherit security from L1, see our blockchain pillar guide.

How do crypto bridges work mechanically?

The two dominant bridging mechanisms:

Lock-and-mint. A user deposits the source asset into a smart contract on the origin chain; the contract locks the deposit; a corresponding amount of a wrapped representation is minted on the destination chain. To reverse, the user burns the wrapped token on the destination chain and the contract releases the locked original. Wrapped Bitcoin (WBTC) on Ethereum is the canonical example: BTC is held in BitGo's custodial reserves; WBTC ERC-20 tokens are minted on Ethereum 1:1 against the reserve. The trust model depends on the custodian.

Burn-and-mint (native). Some assets support native cross-chain transfer without wrapping. Circle's Cross-Chain Transfer Protocol (CCTP) burns USDC on the source chain and mints fresh USDC on the destination chain via a smart contract attestation system. No wrapped token; no custodian holding reserves; the destination USDC is functionally identical to the source USDC. Documentation at Circle's CCTP page.

Liquidity-pool bridges. A user deposits the source asset on one chain; the bridge protocol releases an equivalent from a pre-funded liquidity pool on the destination chain. Across and Stargate are the dominant examples. Faster than lock-and-mint because settlement is instant on the destination side, with the source-side bridge replenishing the destination pool through a separate process.

Messaging-protocol bridges. LayerZero, Wormhole, and Axelar provide general cross-chain messaging that applications use to coordinate state across chains. They are bridge infrastructure rather than end-user products; applications built on top (Stargate on LayerZero, Portal on Wormhole) are what users actually transact with.

What are the main types of bridges?

The 2026 best-practice ordering for material amounts: native-asset routes (CCTP) when the asset supports them; native L2 bridges for L1-to-L2 transfers; mature liquidity-pool bridges (Across, Stargate) for L2-to-L2 and L1-to-non-EVM-L1 transfers; messaging-protocol bridges only through application layers that have been audited and have a multi-month operating history.

How to bridge ETH from mainnet to an L2 (Arbitrum, Base)

Walk through the canonical Ethereum-mainnet-to-Arbitrum route:

1. Hold ETH in a self-custodial wallet (MetaMask, Rabby, Frame). See our wallet pillar guide for the wallet decision tree.
2. Navigate to bridge.arbitrum.io. Connect your wallet on the Ethereum mainnet network.
3. Enter the amount of ETH to bridge. The interface shows the destination Arbitrum address, the gas cost on Ethereum, and the estimated arrival time (typically 10-15 minutes).
4. Approve the bridge transaction in your wallet. The transaction locks ETH in the Arbitrum Bridge contract on Ethereum.
5. Wait for the ETH to arrive on Arbitrum. The wallet network switcher should change to Arbitrum once the funds appear.
6. For the reverse direction (Arbitrum to Ethereum), the canonical bridge has a 7-day fraud-proof window, the L2 withdrawal completes after 7 days. To exit faster, use a liquidity-pool bridge like Across that pays you out immediately on Ethereum at a small fee while the bridge protocol earns the canonical withdrawal in the background.

The same pattern works for Base (bridge.base.org), Optimism (app.optimism.io/bridge), and Polygon zkEVM. Each L2 has its own canonical bridge URL. Always verify the URL through the L2's official documentation or our reviews; phishing bridge sites are a recurring attack vector.

How to bridge USDC across chains using Circle CCTP

Circle's Cross-Chain Transfer Protocol is the cleanest 2026 stablecoin bridge route because it burns and mints native USDC rather than producing a wrapped representation. The destination USDC is identical to the source USDC, no peg risk, no liquidity-pool dependency, no custodial reserves to trust.

1. Visit a CCTP-enabled interface. Direct interface at Circle's CCTP page; integrated interfaces at Bridge.btc.b, Across, Stargate, and many wallet UIs.
2. Connect your wallet on the source chain (Ethereum, Arbitrum, Optimism, Base, Polygon, Solana, Avalanche, and 15+ other supported chains).
3. Enter the amount of USDC and select the destination chain.
4. Approve the burn transaction. The interface displays the gas cost (varies by source chain) and the estimated destination arrival time (10-25 minutes typical).
5. The protocol burns USDC on the source chain and generates an attestation. After Circle's attestation service confirms the burn (a few minutes), fresh USDC mints on the destination chain to your wallet.

CCTP is the recommended route for any USDC cross-chain movement above approximately $1,000. For smaller amounts, the gas cost on Ethereum can make a liquidity-pool bridge (Across, Stargate) cheaper despite the wrapped-token complexity. For stablecoin context including USDC's MiCA registration and reserve structure, see our stablecoin pillar guide.

What are the best crypto bridges in 2026?

• Across. Liquidity-pool bridge with the strongest 2025-2026 security record. Optimistic verification with bond-based dispute resolution. Live TVL on DefiLlama. Recommended default for L2-to-L2 transfers.
• Circle CCTP. Native-asset USDC bridging across 20+ chains. The cleanest stablecoin route. No wrapped token, no liquidity pool, no custodial reserves.
• Native L2 bridges. Arbitrum Bridge, Optimism Gateway, Base Bridge, Polygon zkEVM Bridge. Strongest security model (canonical L1 contracts). Slow on the L2-to-L1 direction (7-day fraud window for optimistic rollups).
• Stargate (LayerZero). Liquidity-pool bridge on LayerZero messaging infrastructure. Broad chain support. Note that the LayerZero stack carried the April 2026 KelpDAO $292M exploit through a misconfigured DVN setup; verify the application layer's DVN configuration before material amounts.
• Wormhole. Messaging protocol used by Portal Bridge and many applications. Higher lifetime processed volume than most competitors; experienced a $326M exploit in February 2022; the protocol has been substantially hardened since.
• Axelar. Messaging protocol with proof-of-stake validator set. Used by Squid Router and many other applications.
• Hop Protocol. Specialized in fast L2-to-L2 transfers with bonded relayers.

The L2Beat security comparison at l2beat.com/bridges tracks trust assumptions and historical incident records for the major bridges.

How much do bridge fees cost?

Bridge cost has three components: gas on the source chain, gas on the destination chain (paid by the bridge in lock-and-mint models, by the user in some liquidity-pool models), and the bridge protocol fee (typically 0.05-0.5% of amount bridged).

Typical 2026 cost estimates:

• Ethereum L1 → Arbitrum via native bridge. $2-15 in Ethereum gas plus near-zero protocol fee. Most expensive component is L1 gas.
• Arbitrum → Base via Across or Stargate. $0.50-2 total. L2-to-L2 routes are dramatically cheaper than anything involving L1.
• USDC cross-chain via Circle CCTP. Source-chain gas ($0.10-$15 depending on chain) plus near-zero protocol fee.
• Arbitrum → Ethereum via canonical (7-day withdrawal). Approximately $1-3 in L2 gas; free protocol fee but capital locked for 7 days.
• Arbitrum → Ethereum via Across (fast). $1-3 L2 gas + Across protocol fee of approximately 0.1% to compensate for the relayer fronting Ethereum-side liquidity.

For frequent users, gas-token holdings on each chain you use ($20-50 worth of ETH on Arbitrum, Base, Optimism each) avoid the slow "bridge gas first, then bridge value" double-step.

What is wrapped Bitcoin (WBTC) and how does it work?

Wrapped Bitcoin (WBTC) is an ERC-20 token on Ethereum that represents Bitcoin 1:1 against a custodial reserve. BitGo Trust Company custodies the underlying BTC; the WBTC DAO authorizes merchants to mint and burn tokens. Approximately 150,000 WBTC circulates as of early 2026, representing approximately $9 billion of BTC locked in BitGo custody.

WBTC exists because Bitcoin does not natively support smart contracts. Wrapping BTC into an ERC-20 token lets BTC holders use their BTC value in Ethereum DeFi (Aave lending, Uniswap LPs, restaking strategies). The trade-off is custodial trust: WBTC holders rely on BitGo to maintain the 1:1 reserve. Alternative wrapped-BTC products include tBTC (Threshold Network's decentralized variant) and renBTC (deprecated post-Ren Protocol shutdown).

The honest framing: WBTC is the dominant 2026 Bitcoin-in-DeFi product but carries custodial risk that pure on-chain BTC does not. For BTC holders who do not need DeFi participation, holding BTC on Bitcoin L1 in a hardware wallet is structurally safer.

How are bridge transactions taxed?

Bridge transactions create tax events that vary by jurisdiction and by bridge mechanism:

• Lock-and-mint bridges. The position changes from the source asset to the wrapped representation. Some tax authorities (IRS guidance has been ambiguous through 2025) treat this as a taxable exchange; others treat the wrapped token as the same asset for cost-basis purposes.
• Burn-and-mint (CCTP). Most jurisdictions treat the destination asset as a continuation of the source basis, no recognition event because the asset is functionally identical.
• Liquidity-pool bridges. Often treated similarly to lock-and-mint, with the exchange rate at the moment of the bridge defining any gain or loss.
• Exchange-mediated bridging. Typically a deposit + withdrawal pair on the exchange; not a taxable event in most jurisdictions because no disposal occurred.

The 2025 US Form 1099-DA reporting requirements for digital-asset brokers have not fully resolved bridge-event treatment as of 2026. Maintain records of every bridge transaction with timestamps, source and destination chains, amounts, and fees. For broader tax context, see our Singapore tax guide and Canada crypto bank guide.

What are the real risks of bridging crypto?

• Bridge exploits. The dominant 2026 risk. Cumulative bridge losses exceed $2.5 billion since 2020. Peckshield tracked 8 bridge exploits through mid-May 2026 with $328.6 million drained cumulatively, including KelpDAO's $292 million LayerZero exploit on 18 April 2026.
• Validator or DVN compromise. Messaging-protocol bridges (LayerZero, Wormhole, Axelar) depend on off-chain validator or DVN sets. Compromise of these external infrastructure pieces produces bridge failure even when the on-chain contracts are correct. The KelpDAO exploit was a 1-of-1 DVN misconfiguration; the underlying LayerZero protocol was not exploited.
• Custodial risk (WBTC, lock-and-mint). Wrapped assets depend on a custodian maintaining reserves. Loss of custody, regulatory action against the custodian, or fraudulent issuance would compromise the wrapped position.
• Liquidity-pool depletion. Liquidity-pool bridges (Across, Stargate) can fail to settle quickly if the destination pool is empty. Funds are not lost but may be stuck pending replenishment.
• Wrong-network errors. Sending ERC-20 USDC to a Solana address (or vice versa) loses the funds. The address format may be valid on the destination chain even though the asset isn't compatible. Always verify the destination network plus address before sending.
• Phishing bridge sites. Fake bridge URLs are a recurring attack vector. Verify URLs through the L2's official documentation or our reviews, not through search-engine ads.
• Slow withdrawal on optimistic rollups. Native L2-to-L1 withdrawals take 7 days due to the fraud-proof window. This is structural, not a bug, but the capital is locked for the period.
• Cross-chain tax recording. Every bridge event needs a record; high-frequency bridge users accumulate dozens of events per year that need reconciliation.

Frequently asked questions

Auto-rendered from the data contract.