Mistral AI for Developers

Try Mistral Vibe: mistral.ai/products/vibe

Packed house at our @mistralvibe hands-on workshop at #AIAgentConference! 🚀

🚀 Unlock AI agent superpowers with @MistralAI at #AgentConference NYC! Join us on May 4 and get hands-on with Mistral Vibe—our terminal-native coding agent. Customize it to accelerate your workflow (yes, even if you’re new to agents!). Bring your laptop, leave with ready-to-use automations. Plus, catch Lélio, our VP Engineering, at the Agentic Engineering track on Tuesday, May 5.

🔌Your MCP Tools - available across every Mistral App Easily register an MCP server for instant availability in Le Chat, AI Studio, and our API.

Watch it in action:

Use our built-in and featured connectors or use your own MCP server. Authenticate and attach it to any conversations or agents. mistral.ai/news/connectors

Build once. Reuse everywhere. The Connectors API is now in Public Preview. Register MCP connectors once and use them across Le Chat, AI Studio, and all programmatic tool calls.

Try it here: console.mistral.ai/build/document…

🎮From Playground to ready runnable Code - Document AI does the work, Studio gives you the code.

🎮Try it here: console.mistral.ai/build/audio/sp…

🎙️Transcribe meetings, interviews, and more - with built-in speaker diarization and segmentation.

🎮Try it here: console.mistral.ai/build/document…

📄 Upload PDF files and extract clean, structured text data.

Try it out: console.mistral.ai/build/document…

Mistral OCR 3 now returns confidence scores, at page level or word level. Know exactly how sure the model is about what it extracted. Available now via API and to test in Mistral Studio.

🎙️Record or upload your voice and generate realistic speech from audio references - controlling tone, rhythm, intonation, and emotional expression.

🎙️Choose your ideal voice and bring your text to life.

👐Learn more about our audio models - and how to build such pipeline - in the blog post below. learn.mistral.ai/home/blogs/des…

⚙️From STT to our Agent and TTS, pick the best model for your use case and build your own real-time, speech-capable system.

🎙️Designing a speech-to-speech assistant Build a speech-to-speech assistant with web search access in 150 lines of code. - Voxtral Transcribe 2 for STT + diarization - Mistral Small 4 for agentic reasoning & efficiency - Voxtral TTS for realistic speech synthesis