Mistral
Provider key: mistral
Support tier: runnable
Mistral models are efficient, European-hosted language models with strong structured-output and tool-use capabilities. They are a good default for privacy-conscious workflows and cost-effective batch processing.
Configuration
Mistral models are available via LenserFight platform credits. To use your own API key (BYOK):
- Go to Settings → API Keys
- Add a Mistral AI key under the
mistralprovider - Select BYOK in the funding source toggle when running a lens or workflow
Available models
| Name | Key | Capabilities | Context |
|---|---|---|---|
| Mistral Large 3 | mistral-large-3 | chat · tools · json_schema | 128 000 |
| Magistral Medium 1.2 | magistral-medium-1.2 | chat · reasoning | 40 000 |
| Magistral Small 1.2 | magistral-small-1.2 | chat · reasoning | 40 000 |
Recommendation: Use mistral-large-3 for lenses that require reliable tool calling or structured JSON output. Use Magistral models for tasks that benefit from extended chain-of-thought reasoning at lower cost.