How to bridge USDC to Circle's Arc: CCTP, Gateway and the steps

Use Circle's own rails. USDC arrives on Circle's Arc through CCTP, which burns USDC on the chain you are leaving and mints native USDC on Arc, or through Gateway, which gives you one USDC balance you can draw on from several chains without a discrete bridge step each time. Both are Circle-built, both are documented for Arc specifically, and neither produces a wrapped token. Third-party bridges may add Arc routes later; as of 6 September 2026 none had published Arc-specific contract addresses or endpoint documentation.

Here is the sequence, then the detail behind each choice.

The steps

1. Decide whether you are moving a balance or building a habit. A one-off transfer is a CCTP job. If you will be pulling USDC onto Arc repeatedly from several chains, Gateway's unified balance is the design that fits, and it is worth the extra integration work up front.
2. Get USDC on a source chain you already use. CCTP moves USDC between chains; it does not create it. If you are starting from fiat, that is an on-ramp problem, covered further down.
3. Confirm the Arc destination values before you touch anything. Chain ID 1243 for mainnet, 5042002 for testnet. Circle had not published the official mainnet RPC or block explorer URLs as of 6 September 2026 — take both from docs.arc.io when they appear, and never from a post, an ad or a search result.
4. Execute the transfer through Circle's own interface or SDK, not a third-party front end you found while searching. The burn on the source chain, the attestation, and the mint on Arc are three steps of one operation; let Circle's tooling sequence them.
5. Wait for attestation, then verify the mint on the official explorer. Arc targets sub-second finality, so inbound CCTP transfers can reach attestation faster than transfers into chains needing several block confirmations. The transfer is not done when the burn confirms; it is done when the mint lands.
6. Fund gas separately, in your head. Gas on Arc is paid in USDC, so the balance you just bridged is also the balance that pays to move it. Bridge more than you intend to spend, and keep a floor in the account — a wallet at zero USDC on Arc cannot transact at all, it does not degrade gracefully.
7. Test with a small amount first, every time you change anything — a new endpoint, a new wallet, a new source chain. The cost of a test transfer is a rounding error against the cost of finding out that a value was wrong.

CCTP, in plain terms

CCTP is Circle's Cross-Chain Transfer Protocol, and Arc's own developer documentation confirms native support with Arc assigned its own CCTP domain. The mechanism is burn-and-mint: the USDC you send is destroyed on the source chain, Circle attests to that, and native USDC is minted for you on Arc.

Why that distinction matters more on Arc than on a general-purpose chain: the older bridge design locks an asset on one side and mints a synthetic representation on the other, which introduces a dependency on the bridge contract staying solvent and uncompromised, and leaves you holding a token that some contracts on the destination chain will not recognise or price correctly. Arc exists to settle stablecoins. USDC that arrives by native mint is the same asset every Arc-native contract expects; a wrapped stand-in would not be. That is a structural reason to prefer the canonical route here, not a preference.

Gateway, which is not a bridge

Circle Gateway is a unified-balance product rather than a cross-chain messaging protocol. You deposit USDC once and can then access that balance's liquidity across supported chains without a separate bridge-and-wait cycle each time you need funds somewhere new. Circle and Arc have published joint developer material on combining CCTP and Gateway to consolidate cross-chain USDC into Arc, including a Unified Balance Kit that exposes the whole thing through one integration.

Pick between them on this basis. CCTP is the right answer when you know where the money needs to be and want it there. Gateway is the right answer when you do not know in advance, and the cost of guessing wrong — capital stranded on the wrong chain — is the thing you are trying to eliminate. Treasury operations and multi-chain applications usually want the second; a person moving a balance wants the first.

Fiat on-ramps are a separate problem

Getting from a bank account or card to USDC is a different piece of plumbing from moving USDC between chains, and it is worth naming as its own step because it is usually the slow one. In practice most routes onto Arc will run through an exchange you already hold an account with, or through a business on-ramp with KYB already completed, and then bridge. Crossmint has announced an Arc partnership covering on-ramps alongside wallet and checkout tooling, without a published launch date as of 6 September 2026. Kraken and Upbit are named in Circle's August 2026 release for USDC access, though exchange-side Arc deposit and withdrawal support has not been independently verified per exchange.

For which bridges and on-ramps are live today, see Arc bridges and on-ramps status, which tracks each provider against whether it has published Arc-specific technical detail rather than a partnership announcement.

The mistake to avoid in launch week

Every chain launch produces a crop of fake bridge front-ends, and Arc's launch has an unusually large opening for them, because the official mainnet endpoint and explorer URLs were still unpublished a week and a half out. A site that looks like a bridge, appears in a search ad, and asks for a token approval is the standard shape of the problem.

Two rules cover almost all of it. Reach Circle's tooling by typing the address, not by clicking a result. And treat any URL claiming to be the Arc mainnet RPC or explorer before you have seen it on docs.arc.io as hostile, because before Circle publishes them there is no way for anyone to have a legitimate one.

What changes after 16 September

CCTP starts moving real value rather than testnet USDC, Gateway's unified balance becomes a live capital-efficiency tool rather than a demonstration, and any third-party bridge claiming to be live on Arc acquires an obligation it does not have today: publishing a mainnet contract address you can check. Until a bridge does that, treat it as part of the conversation rather than part of the infrastructure.