Magistral Medium 2509 is Mistral AI's reasoning model for enterprise use, with 73.6% on AIME 2024 (single-sample), traceable chain-of-thought, and multilingual reasoning across eight languages.
import { streamText } from 'ai'
const result = streamText({ model: 'mistral/magistral-medium', prompt: 'Why is the sky blue?'})Announced March 17, 2025, Magistral Medium 2509 is Mistral AI's first dedicated reasoning model and the more capable of the two Magistral variants. Magistral Medium 2509 scores 73.6% on AIME 2024, a standard mathematical reasoning benchmark, and reaches 90% with majority voting at 64 samples.
Magistral Medium 2509 produces traceable chain-of-thought (CoT) reasoning, exposing each inferential step rather than returning only a final answer. This transparency targets enterprise use cases in regulated industries like finance, legal, and healthcare.
Multilingual reasoning spans English, French, Spanish, German, Italian, Arabic, Russian, and Simplified Chinese. A single deployment can serve global teams without language-specific fallbacks.