Mistral Small and Large
Open-weight LLMs support self-hosted text generation, with function calling in Mistral Large and hosted inference through an API or cloud providers.
Mistral Small and Large are downloadable language models for developers building chat and text applications on their own infrastructure. Their reference library, mistral-inference, is archived and no longer maintained.
- Text generation: Use instruction-following models to produce responses for chat and other text applications.
- Function calling: Mistral Large supports function calls so applications can connect model responses to external tools.
- Self-hosted inference: Downloadable weights and a vLLM deployment option let you serve models on infrastructure you control, including on-premises servers or a VPC.
Mistral also offers hosted inference through its API and cloud providers. Those services process requests on hosted infrastructure rather than in your local deployment. The reference library requires a GPU; its hardware requirements shouldn't be treated as requirements for every serving framework.
The inference code uses Apache 2.0, but model weights have separate licenses. The Mistral Large weights linked by the reference library use the non-commercial Mistral AI Research License. Commercial use depends on the license of the specific weights you choose.
Logotyper
