The Model Index
Models/Mistral

Mistral Medium 3

MistralClosed weightsFlagship

Mistral Medium 3 is a closed AI model from Mistral with an index score of 120 (GPT-4 = 100), rank 156 of 217.

120104–136Index score#156 of 217 models
$0.44/ $2.2Price per million tokensinput / output · via OpenRouter
131KContext windowtokens it reads at once
7 May 2025ReleasedReplaced by Mistral Medium 3.1

Where it sitson the frontier

801001201401601802002023202420252026GPT-4 at launch = 100Mistral Medium 3

o1-mini reached this score 8 months earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence
GPQA Diamond (Epoch AI)
60%
Humanity's Last Exam
5%
Mathsevidence
FrontierMath (tiers 1-3)
no result
Codingno evidence yet
SWE-bench Verified
no result
Terminal-Bench 4.0
no result
Agentsevidence
APEX-Agents
no result
Professional workno evidence yet
Finance Agent (v2)
no result
Novel problemsno evidence yet
ARC-AGI-2
no result
Long tasksno evidence yet
METR time horizon
no result

Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.

What others sayrank on 9 leaderboards

Best attop three among current models

Replaced by Mistral Medium 3.1; it no longer leads any category among current models.

Where to use itprices checked 18 Nov 2026

API providers 1 · per 1M tokens, in / out

Host prices from OpenRouter; going direct can differ.

Facts and sources

Released7 May 2025
Announcementmistral.ai ↗
Release entryMistral Medium 3
Scored asMistral Medium 3
Output speed–
Hugging Faceclosed weights

Mistral releasesaround this one

This page as a chartDownload PNG