Inference

Use Mistral models through the API, compare model options, route requests, and tune generation behavior.

Explore

Find reference material for model choice, pricing, lifecycle, regional routing, and request behavior.

View all models with specifications, benchmarks, and feature comparison.

Compare model pricing by family.

Plan for model maturity stages, aliases, deprecations, and retirement.

Route inference requests to a specific region.

Route eligible API requests through a priority queue.

Explore experimental model releases.

Design prompts for model requests.

Tune generation parameters.

Review endpoint details.

Latest models

A third-party open source text model from Z.ai with a 1M-token context window.

Compact multimodal moderation model for text and image safety classification.

Updated code agent for Lean 4 formal proof engineering and automated theorem proving.

Our latest OCR service with paragraph-level bounding boxes and structural block labels.

Our frontier-class multimodal model optimized for agentic and coding use cases.

State-of-the-art TTS with zero-shot voice cloning and multilingual support.