Mistral AI

1. Platform overview

   The Mistral platform, Studio, API, and how products fit together.

2. Models overview

   Model families for text, vision, code, reasoning, embeddings, OCR, and audio.

3. Model selection guide

   Choosing a model by capability, cost, and latency.

4. First API request and SDKs

   Authenticating and making chat completion calls with the official clients.

5. Chat completions

   Messages, roles, sampling parameters, and streaming in chat completion.

6. Prompting and sampling

   Prompting techniques and the effect of temperature and top_p sampling.

7. Function calling

   Defining tools and letting models call functions.

8. Structured output

   JSON mode and custom schemas for structured responses.

9. Reasoning

   Reasoning models and their thinking output.

10. Embeddings

    Text and code embeddings for retrieval and clustering.

11. OCR and document processing

    Document OCR, annotations, and document question answering.

12. Agents

    Agents and conversations API, built-in tools, and handoffs.

13. Safety and moderation

    Moderating inputs and outputs and guardrails.

14. Batch processing and caching

    Asynchronous batch jobs for large workloads.

15. Usage, limits, and errors

    Rate limits, usage tiers, and error codes.

Each section links to the official page. Titles and summaries by Extra Credit; the docs belong to their authors.