Is Mistral AI Down?

No — Mistral AI is up

Reachable from all 8 checked regions

Average response time: 164ms

Last checked 3 hours ago · checks run every 6 hours

Official status page: https://status.mistral.ai

Mistral AI uptime

100%

Last 7 days

100%

Last 30 days

100%

Last 90 days

141ms

Avg response, 30 days

Measured from multiple regions every 6 hours. Percentages count only checks that returned an availability answer — 13 days measured so far. A dash means that window does not yet hold enough measured days to publish a figure.

30-day history

13-day clean streak

Jul 25 Today

No downtime Partial Downtime Not measurable No data

Reachability by region

Each region runs its own request from a different part of the world. A service can be up for one continent and down for another, which is usually the first sign of a routing or CDN problem.

ewr

104ms

DNS 0ms TCP 3ms TLS 11ms TTFB 77ms

fra

142ms

DNS 0ms TCP 1ms TLS 12ms TTFB 91ms

gru

221ms

DNS 0ms TCP 2ms TLS 10ms TTFB 123ms

iad

68ms

DNS 0ms TCP 1ms TLS 8ms TTFB 51ms

jnb

218ms

DNS 0ms TCP 0ms TLS 7ms TTFB 202ms

lax

166ms

DNS 0ms TCP 1ms TLS 15ms TTFB 100ms

lhr

66ms

DNS 0ms TCP 2ms TLS 10ms TTFB 42ms

nrt

364ms

DNS 0ms TCP 1ms TLS 8ms TTFB 100ms

ord

74ms

DNS 0ms TCP 2ms TLS 8ms TTFB 53ms

What Mistral AI does

Mistral AI builds large language models and serves them through an API and the Le Chat assistant. Its platform covers chat completions, embeddings, OCR, audio, agents, conversations and batch jobs, with a developer console for keys and usage. Applications calling it for generation or embeddings fail in whatever way their own error handling allows.

What an outage looks like

Mistral reports each API surface separately with 90-day uptime figures: chat completions, embeddings, OCR, agents, conversations, audio, integrations, files, batch and workflows, plus the console. A single endpoint can sit well below the others while the overall banner reads operational, so an embeddings or OCR problem is visible only on its own line. Calls typically fail as 5xx errors or time out.

What to do about it

Check status.mistral.ai, which shows per-endpoint uptime over 90 days rather than one verdict, and compare the specific API you call against the rest. Batch and workflows are listed separately from live completions, so a queued job can be affected while interactive calls succeed. Retry with backoff rather than immediately, and check the console for key and quota errors that present like an outage.

Is it down for everyone, or just you?

If this page says Mistral AI is up but it is not loading for you, the problem is between you and them. Run a check against any URL from all 18 regions to find out where it breaks.

Test it yourself

Related services

Mistral AI outage FAQ

The banner says operational but my calls fail. Why?

Mistral reports each API separately with its own uptime figure, and the top-level banner summarises the platform rather than the endpoint you call. Chat completions, embeddings, OCR and audio have visibly different uptime over any 90-day window. Find the line for the specific API in your code path, since a degraded embeddings endpoint is entirely compatible with a green overall banner.

Is my batch job affected if the API is up?

It can be. Batch and workflows are their own components, separate from live chat completions, so queued work can be delayed or failing while interactive requests succeed normally. If a batch job is overdue and interactive calls work, check those specific lines before assuming your submission was malformed. The reverse also happens during incidents that affect only the synchronous path.

Should my application retry when Mistral returns an error?

Retry with backoff rather than immediately. Tight retry loops against a degraded endpoint add load without improving your odds, and they can turn a partial degradation into a longer one. Distinguish 5xx responses, which are worth retrying, from key, quota and rate-limit errors, which are not and which look like outages in logs that only record that a call failed.

Does an API incident affect Le Chat?

Not necessarily. The assistant and the developer APIs are tracked as separate services on the status page, so Le Chat can be usable while an API endpoint is degraded, or the other way round. If you are checking on behalf of an application rather than your own use, test the specific API rather than concluding from whether the chat interface responds.

How we measure this

We request Mistral AI's public endpoint every 6 hours from Fly.io regions across six continents — 8 of them answered the most recent check.
A region counts as down only when it gets no usable HTTP response. A 403 or 429 means the origin answered and refused us, which we report as blocked, never as an outage.
A single failing region is treated as probe noise. We only change the verdict when two consecutive cycles agree.
Response times average only the regions that actually served the page, so a timeout never inflates the number.
Where Mistral AI publishes an official status feed we read it too. An all-clear from the vendor can soften an unconfirmed degradation; a vendor-declared outage only worsens our verdict when our own checks corroborate it.

Get alerted when Mistral AI goes down

This page refreshes every 6 hours. Your own monitors run as often as every 30 seconds, from the same 18 regions, and tell you the moment something breaks.

Start Free Monitoring

Free plan available No credit card required