MKT
Centralised exchange

BingX

A $150m fund genuinely scoped to theft — with no address, no proof of funds and no published claims process. Its own hack compensation is unconfirmed.

Founded
2018
Registered
Singapore

The short answer

Its Shield Fund is framed as real theft cover rather than the derivatives backstop most venues pass off as insurance, which is to its credit. The catch is that its $150m size and composition are entirely self-asserted: no on-chain address, no proof of funds, no published claims process. That matters more than usual because BingX has already had a $43–52m hack, and the compensation it pledged has never been independently confirmed. We score custody 4 out of 10.

Key facts

Shield Fund$150m, framed as genuine incident and theft cover rather than a liquidation backstop — and entirely self-asserted: no on-chain address, no proof of funds, no published claims processas of 14 Jul 2026
Hack$43–52m, with a compensation pledge that nobody has independently confirmedas of 14 Jul 2026
Proof of reservesMonthly Merkle-tree proof covering liabilities, with an open-source verification tool — genuinely better machinery than most of its tier. No independent signature on it.as of 14 Jul 2026
Custody score4 out of 10as of 14 Jul 2026
Effective cost, $10,000 BTC buy (measured by us)0.1032%; 0.1061% at $100,000as of 14 Jul 2026
Order-book depth within 0.1% (measured by us)109.4 BTC — the third deepest we measuredas of 14 Jul 2026
API response latency (measured by us)460 msas of 14 Jul 2026

A fund with the right scope and no evidence behind it

BingX gets the framing right where most of this category gets it wrong. Its $150m Shield Fund is presented as cover for incidents and theft, not as a derivatives liquidation backstop — which is what the "insurance funds" at Bitfinex, Phemex and Gate.io turn out to be when you read their own documentation. Scoping a fund to the thing users actually fear is the first requirement, and BingX meets it.

It then fails the second. The size and composition are entirely self-asserted. There is no on-chain address, no proof of funds, and no published claims process — so you cannot confirm the money exists, and you cannot find out in advance what would qualify you to claim against it. Compare MEXC, which publishes an address for its $100m fund and lets you check.

That gap is not hypothetical here. BingX lost between $43m and $52m in a hack and pledged compensation, and that pledge has never been independently confirmed. The one occasion where the fund’s existence could have been demonstrated passed without a verifiable demonstration.

The reserve machinery is better than the fund: a monthly Merkle-tree proof covering liabilities with an open-source verification tool, so you can confirm your own balance is in the tree. What it lacks, again, is an independent signature. We score custody 4 out of 10.

Cheap, and deeper than it looks

A $10,000 bitcoin buy cost 0.1032% on 14 July 2026, in line with the cheapest tier, rising to 0.1061% at $100,000.

The depth is the surprise: 109.4 BTC within 0.1% of best ask, the third deepest book we measured, behind only Kraken and Binance and well ahead of better-known venues. For execution at size this is a genuinely strong venue, and the flat cost curve reflects it.

API response latency measured 460 ms, in line with Binance and KuCoin.

As with most of this tier, the execution is good and the custody evidence is the constraint. The pattern repeats often enough across our rating to be worth stating as a rule: how well a venue trades tells you almost nothing about whether it can prove it holds your money.

Incident record

DateTypeAmountWere users made whole?
20 Sept 2024Hack or exploit$52,000,000A compensation pledge was made and has never been independently confirmed. The loss is reported between $43m and $52m.source

In our ratings

Compared with

Our coverage of BingX

Questions people ask

Is BingX safe?

Custody scores 4 out of 10. Its Shield Fund is correctly scoped to theft rather than derivatives liquidations, and its $150m size is entirely self-asserted with no address, no proof of funds and no published claims process. Its own $43–52m hack produced an unconfirmed compensation pledge.

What is the BingX Shield Fund?

A stated $150m fund for incidents and theft — the right scope, unlike most exchange "insurance", which turns out to be a derivatives backstop. What is missing is any evidence: no on-chain address, no proof of funds, no published claims process.

Were BingX users compensated after the hack?

A pledge was made and has never been independently confirmed. The loss is reported at between $43m and $52m. It was the one occasion the fund could have been demonstrated, and it was not demonstrated verifiably.

Is BingX liquid?

More than its profile suggests — 109.4 BTC of depth within 0.1% of best ask, the third deepest book we measured, behind only Kraken and Binance. A $10,000 buy cost 0.1032% and a $100,000 one 0.1061%.

What changed

  • 27 Sept 2026 — Profile published.