Skip to main content
mistral.health
Mistral

Is Mistral down? Live API reachability checks

Mistral AI API and console connectivity

Check whether selected Mistral API and console addresses are reachable from your browser, then compare with official incident reports.

Quick answer: These browser checks show whether selected public addresses respond from your connection. They do not inspect HTTP status or response bodies, test an authenticated model request, or confirm that chat, embeddings or Codestral are working. A failed check can reflect your network or browser restrictions rather than a service outage. Check the official Mistral status page for reported incidents.

Mistral API pricing & cost calculatorCompare models using your request volume, tokens and cache share.

Mistral API Status

Live, in-browser checks against Mistral's public APIs. Refreshes every 15s.

Checking…

Mistral API

Whether selected api.mistral.ai addresses respond from your browser

3 endpoints
  • Models
    GET /v1/models
    —ms
    warming up
  • Chat Completions
    POST /v1/chat/completions edge
    —ms
    warming up
  • Embeddings
    POST /v1/embeddings edge
    —ms
    warming up

Console & marketing

Whether the console and website respond from your browser

2 endpoints
  • Console
    console.mistral.ai
    —ms
    warming up
  • Marketing site
    mistral.ai root
    —ms
    warming up
Polling every 15s · Last updated — · Next in 15sOpen Mistral ↗

Your browser can only tell whether each address responded; it cannot see HTTP status codes or response bodies. A failed check may come from your network or browser settings. For reported incidents, see the official status page below.

Mistral API reachability

The live check shows whether api.mistral.ai/v1/models, /v1/chat/completions and /v1/embeddings, plus the console at console.mistral.ai, respond from your browser. These addresses belong to Mistral's hosted API platform, called La Plateforme.

Le Chat status

Le Chat is the user-facing chat service at chat.mistral.ai; the widget does not directly check that page or chat replies, so compare any symptoms with the official status page.

mistral.ai website status

The marketing-site check covers reachability of the homepage at mistral.ai, not model requests or Le Chat replies.

How we check

While this page is open, your browser checks api.mistral.ai/v1/models, api.mistral.ai/v1/chat/completions, api.mistral.ai/v1/embeddings, console.mistral.ai and mistral.ai on load and every 15 seconds. Each check has a six-second timeout. These are lightweight reachability checks, not authenticated model requests: they do not generate chat replies or embeddings.

The widget shows “Service Disruption” when any latest check fails or times out, and “All Systems Operational” when all checked addresses are reachable. A failed check means that address could not be reached from your browser; it does not by itself confirm a provider-wide outage.

When diagnosing an outage, repeated HTTP 5xx server errors or timeouts can indicate a service is down; intermittent failures, elevated error rates or slower responses can indicate degraded service. Your browser cannot see HTTP status codes or response bodies from these checks, so the widget cannot detect HTTP 5xx errors, measure error rates or assign a separate degraded verdict. A reachable address can still have an application-level problem.

Checks run from the visitor's browser, not a central monitoring server. A corporate network, VPN, firewall or local connection problem can produce a false “down”. Try another network or turn off your VPN, compare with status.mistral.ai, and check your account's rate limits before treating a failed probe as an outage.

Sponsored · Mistral
Ask Mira
La Plateforme · EU

Hi, I'm Mira — your Mistral guide. Ask about models, pricing, rate limits, or function calling.

Open La Plateforme

Sponsored content. mistral.health may earn a referral reward when you sign up via this widget. Answers are informational only.

Official Mistral status page

Open ↗

About Mistral

Mistral AI is a Paris-based frontier lab shipping open-weight and commercial LLMs (Mistral Large, Small, Codestral, Ministral) through the Mistral API at api.mistral.ai. This page checks whether selected API addresses respond from your browser; it does not send authenticated model requests.

Mistral at a glance

Founded
2023
HQ
Paris, France
Flagship models
Mistral Large, Codestral
Open weights
Yes (Apache-2.0 for some)
API base
api.mistral.ai
Docs
docs.mistral.ai

Common Mistral issues

  • 429 rate-limit spikes

    The Mistral API enforces per-model TPS and TPM caps. Bursty traffic on Large hits limits before Small does — check the Retry-After header and back off exponentially.

  • Regional latency to EU-only endpoints

    Mistral serves inference from EU regions by default. Callers in APAC or the Americas can see 200–400ms baseline latency on top of decode time.

  • Function-calling schema errors

    Tool JSON schemas must be strict Draft-07 with additionalProperties: false. A loose schema returns a 400 the docs describe as invalid_tool_schema.

Background

Mistral AI was founded in April 2023 by former Meta and DeepMind researchers and released the first open-weight Mistral 7B that autumn. Mistral's hosted API launched in December 2023; the commercial Mistral Large and code-specialist Codestral rounded out the API through 2024–2025.

Frequently asked

Is the Mistral API down right now?
The dashboard above checks selected api.mistral.ai addresses from your browser every 15 seconds. Green means the address responded from your connection, not that model requests are working; check status.mistral.ai for reported incidents.
Which models are on the Mistral API?
Mistral Large, Mistral Small, Codestral, Ministral, and the embed model — plus fine-tunes you create. Availability varies by region and tier.
Where can I see rate limits?
Rate limits are per API key and per model tier; the current caps live in your console.mistral.ai dashboard under Usage.
How can I check if Mistral AI is down?
Use the browser reachability checks as a first signal and compare with official incident reports. For a specific API problem, inspect the response to the actual operation in your own authenticated application. A successful model-list request only establishes that listing worked; it does not confirm other endpoints.
What do common Mistral API HTTP error codes mean?
A 401 error indicates an invalid API key, 429 signifies rate limiting or insufficient credit, and 500 or 503 errors represent internal server or provider-side capacity outages.
Where are Mistral AI status incidents officially published?
Mistral AI posts official incident reports and platform maintenance notices on status.mistral.ai, while mistral.health shows whether selected addresses respond from your browser.
What is the base URL for the Mistral API?
The official base URL for Mistral API requests is https://api.mistral.ai/v1. It powers all chat, embedding, and code completions endpoints.
How do I check if the Mistral API is working?
The checks on this page only show whether addresses respond from your browser. For a specific operation, send that request from your own authenticated application and inspect the response body and status for that call. A successful response confirms that operation, not the whole account-wide API; compare with status.mistral.ai for reported incidents.
Is Mistral down right now?
Check the live reachability results at the top of this page, which show whether selected Mistral addresses respond from your browser, and compare with the official status page.
Does Mistral AI have an official status page?
Yes, status.mistral.ai is the official status page. This site is an independent checker that shows whether selected addresses respond from your browser.
Why does the Mistral API return 429 errors?
429 is rate limiting, not an outage: your account exceeded its requests-per-minute or tokens-per-minute allowance. Lower concurrency, add backoff, or request a limit increase in the Mistral console.

About this check

All data is provided on a best-effort basis with no guarantees about accuracy or availability. This is an unofficial status page and is not affiliated with, endorsed by, or sponsored by Mistral.

Checks run client-side from your browser against Mistral's public endpoints. Latency reflects your network path, not the server's health alone. See our resources for more on how to read these checks.

Related on mistral.health

All tracked APIs

Live checks across frontier LLM APIs and the open ML hub.

LLM APIs

Mistral

Mistral AI API and console connectivity

View status →
DeepSeek

Low-cost frontier LLM API (V3 + R1)

View status →

ML hub

HuggingFace

The open ML hub · models, datasets, Spaces

View status →