About Mistral AI
Mistral AI builds large language models and offers many of them as open-weight models. Customers can use them as they are or adapt them to their own data and rules.
What it does
The platform covers base models, a tool for customizing them, a studio for building agents and applications, and several hosting options. Models can run on your own hardware, in your private cloud or in Mistral's cloud. That flexibility suits teams with strict requirements about where their data is processed.
Key features
- Open-weight models you can keep and run
- Customization around your own data
- Studio for building and governing agents
- On-premise, private cloud or hosted options
- Dedicated GPU capacity for steady throughput
Who it's for
- Enterprises with data residency requirements
- Public sector and regulated industries
- Developers who want to build custom models
Pricing and good to know
Plans and rates depend on the deployment option, from your own servers to Mistral-hosted inference. Check the official site for current pricing before you commit to a setup. Running models on your own infrastructure means you also manage the hardware, updates and security.
Highlights
- Open-weight models you can keep and run
- Tune models on your own data
- Build and govern agents in one studio
- Run on your servers, a private cloud or hosted
- Reserve GPU capacity for steady workloads
