Le Chat is the assistant built by Mistral AI, a French company that publishes several of its models under open licences and offers European data hosting. The assistant covers general questions, document analysis, image generation, web search with citations, code assistance and a canvas for iterating on documents. Connectors link it to common workplace tools, and agents can be configured for repeated tasks. Its appeal is partly practical, since responses are notably fast, and partly regulatory, as organisations subject to European data rules can use it with clearer guarantees than the American alternatives offer. It suits European businesses, privacy-conscious users and developers who want open weights with a commercial platform behind them. The limitations are ecosystem and depth: it has fewer integrations and third-party extensions than the largest assistants, and on the hardest reasoning tasks it generally trails the leading models.
How to use Le Chat
1Sign up at chat.mistral.ai with an email account; a free tier covers everyday use and paid plans raise limits and add features.
2The interface will feel familiar immediately, so the setup worth doing is under the hood: enable the connectors for the services you use, and check the data residency and training settings, since those are the reasons many people choose this assistant in the first place.
3Web search returns citations, which are worth following for anything factual.
4The canvas is useful for drafting a document you keep revising rather than regenerating.
5For developers, the same models are available through the Mistral platform API with straightforward per-token pricing, and several model families can be downloaded and run locally.
6Two things to note: the assistant is faster but less exhaustive than the largest competitors, so complex multi-step reasoning may need more guidance from you, and the library of community-made extensions is much smaller.
Best for
European data requirements
fast everyday answers
document analysis
developer access to open models
privacy-conscious teams
Pros and cons
What it does well
European hosting and clearer data handling options.
Several underlying models published openly.
Notably fast responses.
Web search returns citations.
Where it falls short
Trails the leading models on hard reasoning tasks.
Fewer integrations and third-party extensions.
Smaller community and less shared prompting material.
Le Chat pricing
Free tierPrices as listed by the maker; check the site for the latest.
Free tier available
paid plans and usage-based API pricing, with European hosting options
general assistantopen weightseuropean hostingdata privacyfast inference