Best Cross-Chain Bridges in 2026

How cross-chain bridges work

A bridge connects two or more blockchains through contracts, validators, messaging infrastructure, or liquidity providers. In a lock-and-mint design, an asset is deposited into a contract on the source chain and a wrapped representation is created on the destination chain. Returning the asset burns or releases that representation. The wrapped token depends on the backing asset and bridge contract.

A liquidity network pays the recipient from funds already held on the destination chain. The sender's asset is then balanced across pools through a fee or rebalancing process. Intent-based systems let a sender specify the desired outcome while solvers compete to complete it. Messaging protocols pass verified instructions between chains, allowing an application to control deposits, minting, or transfers.

Aggregators route transfers through one or more bridge systems. They expose more routes, but route selection adds another contract or service layer. Canonical bridges are created or endorsed by a network and often have narrower chain coverage.

How to choose a bridge

Bridge type

The design should match the asset and destination. Lock-and-mint bridges need evidence that wrapped assets are backed and redemption works. Liquidity networks need usable pools on both sides. Intent-based routes need transparent solver terms and settlement rules. Messaging protocols need clear application configuration, since the application may select the verifiers that approve a message.

A weak offer hides the route or changes it without clear notice. It may present a messaging protocol as a complete asset bridge even though the application controls the final transfer logic. LayerZero is an example of a messaging protocol with application-configured decentralized verifier networks. Wormhole is another messaging protocol, using a Guardian network of 19 validators.

Chains supported

Chain count is useful only when the required route is supported and the asset exists in the required form. Check the exact source and destination networks, token contract addresses, minimum deposits, settlement conditions, and whether the route is native or wrapped. A bridge can support many chains while offering few direct routes.

LayerZero states support for 132 blockchains. Wormhole supports more than 45 chains and is used for Portal and Native Token Transfers. The route and transfer method matter more than a headline chain count.

Fees

The full cost can include source-chain gas, a bridge fee, destination-chain gas, a liquidity-provider charge, and a price difference in the route. Some services deduct the charge from the sent amount; others quote it separately. A low displayed fee can still produce a costly transfer on a congested source chain.

LayerZero charges a messaging fee plus source-chain gas. Its native USDT0 transfers carry no USDT0 fee, although network gas and other route conditions still matter. Fees can vary as gas prices, liquidity, or solver quotes change.

Typical transfer time

Transfer time depends on source-chain confirmation, message verification, liquidity, and destination execution. A route that appears instant may still wait for finality or require a manual claim. Check the stated time for the exact asset and route, along with the status shown when a transfer fails.

A weak bridge gives one broad speed claim without separating initiation from verification and destination settlement. The practical comparison is the complete time to usable funds, not the time until a transaction appears on the source chain.

Security model

Security comes from the mechanism that approves a cross-chain message. Multisig bridges rely on a group of signers, so a threshold compromise can authorize a false transfer. Validator networks distribute that authority, but the number of validators, independence, signing threshold, key custody, and upgrade powers still matter. Light clients verify another chain's consensus proofs directly. Optimistic systems accept messages unless a challenge is submitted during a dispute window.

Application-configured verifier networks create a further dependency: the application selects the verification configuration. Look for documented validator or verifier identities, threshold rules, emergency controls, upgrade permissions, and a response plan for invalid messages. A small or opaque signer group is a weak point.

Audits and past exploits

An audit can identify code defects and access-control errors, but it covers a defined version and scope. Check the audited contracts, audit date, unresolved findings, and later upgrades. Past incidents need the exploit date, affected contract, attack path, amount lost, pause procedure, and repayment or recovery terms.

Both LayerZero and Wormhole are marked as audited in this list. LayerZero's summary also refers to a KelpDAO exploit, which exposed the risk of weak verifier configurations. The incident history of a bridge and of the applications built on it are separate checks.

Token and transfer form

A bridge token may govern the protocol, pay fees, or have no role in a particular transfer. Check whether the destination asset is issued by the destination chain, backed by locked collateral, or represented by a wrapped contract with its own redemption process.

Costs and fees to watch

Bridge costs usually combine network gas with protocol or route fees. Liquidity routes can add a spread, while intent systems may include solver compensation. A failed transaction can still consume source-chain gas, and a destination claim may create another fee.

Fee quotes should identify the asset charged, the network used for payment, and whether the destination amount is fixed. Native USDT0 transfers through LayerZero have no USDT0 fee, while the protocol still charges a messaging fee plus source-chain gas. A zero fee for one token or route does not mean the whole transfer is free.

Risks and red flags

Bridge hacks

Bridge contracts hold or control concentrated value, which makes access-control bugs attractive targets. Check the contract addresses, pause controls, upgrade authority, audit scope, and incident record. A past exploit with an unclear affected amount or recovery process leaves the economic exposure hard to judge.

Multisig or validator compromise

Signer keys, validators, Guardians, and verifier networks can approve false messages if attackers reach the required threshold. Check the threshold, key storage, and validator independence. Also check whether one operator controls several identities. A named group is not the same as independent control.

Wrapped-asset depegs

A wrapped token can trade below its intended value if backing is missing, redemption stops, or route demand falls. Check the backing contract and supply and redemption records. Issuer permissions and the exact destination-chain token contract also matter.

Stuck transfers

Transfers can remain pending after a source transaction confirms because verification or liquidity failed. Destination execution can fail too. Check the transfer status tool, retry and refund rules, claim requirements, support channel, and the time after which recovery begins.

Liquidity shortfalls

Liquidity networks may not have enough destination funds for a large transfer. The result can be a delay, a partial route, a high spread, or a rejection. Check pool depth for the exact asset, transfer limits, solver availability, and whether the quote can expire before settlement.

Availability, KYC, and regulation

Many on-chain bridges let a wallet initiate a transfer without an account or identity check, while a hosted interface, fiat on-ramp, card service, or institutional route may apply KYC rules. KYC status belongs to the specific interface, not only to the underlying protocol.

Availability varies by source chain, destination chain, token, wallet, and jurisdiction. Sanctions controls, blocked addresses, and regional restrictions can prevent a route even when both chains appear supported. In the European Union, services that fall within the relevant crypto-asset rules may face obligations under MiCA, while the treatment of a permissionless protocol can differ from that of a regulated interface.

Bridge transfers can create taxable disposal, exchange, or income events depending on the jurisdiction and the asset's legal treatment. Rewards, rebates, liquidity-provider payments, and governance-token distributions may have separate reporting treatment. A bridge's token is not automatically a transfer fee or a reward.