Mistral Large 3
Mistral Large 3 is an open-weights AI model from Mistral with an index score of 117 (GPT-4 = 100), rank 162 of 217.
11799–135Index score#162 of 217 models
$0.50/ $1.5Price per million tokensinput / output · via OpenRouter
262KContext windowtokens it reads at once
2 Dec 2025ReleasedReplaced by Mistral Large 4
Where it sitson the frontier
o1-mini reached this score 15 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsevidence
Professional workevidence
Novel problemsno evidence yet
Long tasksno evidence yet
Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.
What others sayrank on 9 leaderboards
The Model Indexthird-party tests, our method#162of 217 · 117
LMArena · Codingcoding questions in chat#109of 357 · 1471LMArena · Textoverall chat quality, judged by people#116of 362 · 1415LMArena · Instruction followingfollowing detailed instructions#120of 362 · 1403LMArena · Creative writingcreative writing#122of 360 · 1375LMArena · Long promptslong, complex requests#119of 340 · 1419LMArena · Mathsmaths questions in chat#133of 346 · 1402LMArena · Visionunderstanding images#79of 145 · 1200LMArena · WebDevbuilding web apps and front ends#125of 133 · 1231Best attop three among current models
Replaced by Mistral Large 4; 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
Released2 Dec 2025
Announcementmistral.ai ↗
Release entryMistral Large 3 (+ Ministral 3)
Scored asMistral Large 3
Output speed–
Hugging Facemistralai/Mistral-Large-3-675B-Instruct-2512 ↗