Mistral
Mistral chat, embeddings, transcription and document OCR.
Use Mistral models through DeepintShield with your own Mistral API key, including the OCR endpoint for documents.
What works
- Chat Completions, and the Responses API through chat
- Embeddings and list models
- Transcription (STT) with SSE streaming
- Document OCR through
POST /v1/ocr - Tools with auto, any or none choice, and parallel tool calls
Limits to know: No text completions, speech, files or batch. Image generation is not yet implemented in the DeepintShield integration. Forcing one specific tool is not supported.
How to connect
Add your Mistral API key in Model Hub → AI Providers, create a Virtual Key, then call mistral/<model> through https://app.deepintshield.com/v1 with any OpenAI-compatible client.
Read the Mistral provider docs. Requests get the same Virtual Key budgets, guardrails and logs as every other provider on the AI gateway.