Axelar Review

Axelar scores 8.6 out of 10 in our bridge ratings, unchanged from July and the highest mark in the category on current scores, shared with Wormhole. Under the site's alphabetical tie rule Axelar is listed first of nine. It takes full marks for its licence and its audit archive, and near-full marks for its trust model and chain coverage. The one criterion that holds it back is cost: Axelar explains how its fee is built but publishes no figures a user could compare before requesting a quote.

This assessment is desk research from Axelar's documentation, GitHub repositories and blog, rechecked on 6 November 2026. We did not send a transfer through Axelar, and nothing here measures speed or the price of a real transfer.

How Axelar scores in bridges

The bridges category uses five criteria, each worth 20%. All five marks held on this check.

Arithmetic: (5 + 9 + 10 + 9 + 10) / 5 = 43 / 5 = 8.6.

What changed: no mark moved. The audit archive has grown from the "over forty" reports recorded in July to 61, which confirms the existing 10 rather than raising it. The chain-coverage finding is now backed by Axelar's own deployment file as well as its marketing claim.

Strongest criteria: audits and licence

Audit disclosure (10)

Axelar keeps a public audit repository on GitHub. When the company profile was checked on 1 November 2026 it held 61 reports dated from November 2021 to August 2026. Most are from Ackee Blockchain, with others from NCC Group, OtterSec, Halborn, Certik, Cure53, Informal Systems, Least Authority, Common Prefix and FYEO. Every report is a file a reader can open, it is dated, and the firm is named. That is the full standard for this criterion. An audit reviews code at a point in time; it is not a guarantee against a future exploit.

Licence & code openness (10)

The core node software, axelar-core, carries the Apache License 2.0, a permissive open-source licence. Anyone can read, fork and reuse it without a change date or use restriction.

Trust model disclosure (9)

The documentation explains who can move funds and under what threshold. Validators run nodes for each connected chain and vote on whether a transaction happened. Gateway contracts on each chain are controlled by a key split into shares, with each validator's share count set by the AXL staked with it, and a threshold of shares is needed to sign. The homepage claims 75+ validators. The structural risk follows directly from that description: if enough stake-weighted shares were compromised or colluded, the gateways could approve transfers that never happened. The mark stops one short of full because the documentation does not state the signing threshold as plainly as Across states its one-honest-disputer condition.

Chain coverage (9)

Axelar claims more than 80 blockchains for its token service. The public mainnet deployment file lists gateway configurations for 31 chains outside the Cosmos ecosystem, including 27 EVM chains as well as Solana, Sui, Stellar and the XRP Ledger. Cosmos chains connect separately. Chains can also be dropped by governance vote: a May 2026 proposal sought to end the Flow, Berachain and Plume integrations. Check Axelarscan before sending to a less common network.

Weakest criterion: fee transparency (5)

The gas service pricing page describes up to four components, all prepaid in the source chain's native token. A base fee depends on the chain pair. An execution fee depends on the gas limit and destination gas price. An L1 data fee applies to rollup destinations, and an express fee applies only to the Express relayer. Unused gas is refunded on the source chain, and the page gives the refund formula.

The page publishes no amounts, rates or sample costs. The only way to see a number is to request an estimate through the SDK or look up a past transaction. This is a measured deficiency, not a gap in our research: the structure is clear, but the per-pair base fee schedule is not shown. Apps that route through Axelar may add their own fee on top, which the network's documentation does not cover. Our guide to how gas works explains why the execution part moves.

What the score does not cover

• Actual cost and settlement time. Not measured.
• Operator changes. On 15 December 2025 Axelar's blog announced that Circle had agreed to acquire the Interop Labs team and its proprietary intellectual property, with the network, the Foundation and AXL staying independent and Common Prefix taking over many of Interop Labs' activities. We found no closing announcement on Axelar's blog. The bridge criteria do not measure who develops a protocol, so this did not move a mark.
• Incident history. The homepage claims "zero exploits." That is Axelar's statement, and it is not scored.
• AXL staking. Token economics and staking rewards are outside a bridge rating.

Axelar vs Celer cBridge and Across

On current category marks, Axelar and Wormhole share the lead at 8.6, Celer cBridge is third of nine at 8.4, and Across is sixth at 7.2. Axelar and Celer are level on licence, trust model and chain coverage; Axelar's dated 61-report archive outscores Celer's five undated reports (10 against 7), while Celer is ahead on fees (7 against 5). Axelar beats Across on licence (10 against 4) and audits (10 against 8); the two are level on fees at 5 and chain coverage at 9, and Across is a point ahead on trust model disclosure.

Who it suits

Axelar's marks favor developers and users who weigh open code and a deep, dated audit record most heavily, and who accept that the cost of a transfer is known only after asking for a quote. Most people meet Axelar indirectly, through a wallet or app that routes over it, so the app's own fees also matter. For the trade-offs between bridge designs, see how blockchain bridges work and why they are risky. A high score here describes disclosure quality; it is not a promise that funds are safe.