Fireblocks on Arc, Circle's stablecoin chain
Custody and wallet support decide whether a chain is usable. Of the three names attached to that job on Circle's Arc, only one has a joint announcement behind it.
Fireblocks, Ledger and MetaMask are named in Circle's 5 August 2026 press release among the firms enabling access to USDC on Circle's Arc, but the evidence behind the three is not equivalent. Fireblocks has a separate joint press release with Circle stating that its "early support for Arc" will let institutions build, deploy and transact on the chain. Ledger and MetaMask have a name in a list and nothing further that could be found — no supported-network page, no release notes, no changelog entry. Ledger and MetaMask should be read as announced.
Built on Arc (builtonarc.app), an independent directory of what is building on Circle's Arc blockchain, listed [N] projects on [DATE] and marked [N] as live on mainnet.
Fireblocks: the one with a document
Fireblocks is a digital asset infrastructure provider used by exchanges, banks and fintechs for key management, transaction policy and settlement. Its customer base is the population Arc is aimed at. The Circle–Fireblocks release covers more than chain support: it describes interoperability between the Fireblocks Network and the Circle Payments Network, framed as a unified experience for institutions spanning cross-border, merchant and retail treasury use cases. Fireblocks also appeared in Circle's 28 October 2025 public testnet release among the participating firms.
"Early support" is doing real work in that sentence and should not be over-read. It signals that Arc is on the roadmap and probably in a staging environment; it is not a statement that any customer can select Arc in production today, and no date is attached. Still, a joint release from both parties is a stronger source than an ecosystem list, and it is the strongest evidence any wallet or custody name on this chain currently has. The entry for Fireblocks on Built on Arc records that source.
Ledger and MetaMask: what support would even mean
The two are usually grouped and shouldn't be, because supporting a new EVM chain costs them very different amounts.
MetaMask is software, and adding an EVM network to it is close to trivial — a chain ID, an RPC endpoint, a currency symbol, an explorer URL. Any user can add a network by hand today; the question is whether MetaMask ships Arc as a default, discoverable network with a curated RPC, which is an editorial and infrastructure decision rather than an engineering one. Note that the practical obstacle is Circle's: official mainnet RPC and explorer URLs had not been published as of early September 2026, and a wallet cannot ship a default network without an endpoint.
Ledger is hardware, and hardware support is genuinely different. A Ledger device signs transactions in a secure element and displays what it is signing; supporting a chain properly means the device can parse and render the transaction so the user is not blind-signing. For an EVM chain this is usually handled by the existing Ethereum application, so basic signing may work from day one — but that is not the same as Ledger Live listing Arc, showing balances, or the device rendering USDC gas correctly in a fee prompt. Anyone treating a hardware wallet as Arc-ready should test with a trivial amount. The directory entries for Ledger and MetaMask hold whatever evidence exists for each; at the time of writing that is Circle's release.
The USDC gas problem nobody has written about
Arc pays gas in USDC. That is the chain's most distinctive property and it is also the thing most likely to break a wallet's assumptions. Standard EVM tooling models a native coin with eighteen decimals used for fees and a separate ERC-20 for the dollar. Arc collapses those into one asset with six decimals, which touches fee estimation, balance display, "insufficient funds" logic, and every place a wallet reserves a native balance so the user cannot spend themselves into a state where they cannot transact.
None of that is hard, but none of it is free, and it explains cautious wallet language better than reluctance does. It is also why a self-custody user should not assume a manually added network behaves correctly just because transactions succeed.
What is firm about the chain
Mainnet 16 September 2026; public testnet since 28 October 2025; EVM execution; Malachite Tendermint-family BFT consensus with sub-second finality; a permissioned validator set; mainnet chain ID 1243 against testnet 5042002 per ChainList. Circle had not published official mainnet RPC or explorer URLs as of early September 2026 — which is the single fact most likely to be out of date by the time this is read, and the first thing to check.
Why it matters
A settlement chain with institutional validators and no custody support is a demonstration. Fireblocks is the name that converts Arc from a Circle product into something a bank's operations team can hold, because policy engines and transaction controls are what compliance signs off on, not block times. MetaMask is what makes the chain visible to the developers who build on top of it. Ledger is what lets a treasury hold keys itself rather than through a provider.
Three unlocks, one confirmed source between them. That ratio is the honest summary of where Arc's wallet and custody layer stood in launch week, and the thing to re-check rather than assume.
