Mistral Large 4
Mistral Large 4 is an open-weights AI model from Mistral with an index score of 160 (GPT-4 = 100), rank 48 of 207.
160144–177Index score#48 of 207 models
$1.36/ $4.18Price per million tokensinput / output · via OpenRouter
1.05MContext windowtokens it reads at once
6 Oct 2026Released21 days ago
Where it sitson the frontier
Gemini 3 Pro reached this score 11 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsevidence
Professional workevidence
Novel problemsevidence
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 12 leaderboards
The Model Indexthird-party tests, our method#48of 207 · 160
Artificial Analysis Intelligence Indexoverall capability across ten evaluations#29of 216 · 38LMArena · Mathsmaths questions in chat#51of 346 · 1467LMArena · Codingcoding questions in chat#72of 354 · 1498LMArena · Instruction followingfollowing detailed instructions#85of 359 · 1426Epoch Capabilities Indexoverall capability from many benchmarks#59of 243 · 150.0LMArena · Textoverall chat quality, judged by people#91of 359 · 1434LMArena · Long promptslong, complex requests#98of 337 · 1436LMArena · Visionunderstanding images#44of 145 · 1252LMArena · WebDevbuilding web apps and front ends#41of 126 · 1545LMArena · Creative writingcreative writing#124of 357 · 1369LiveBench · Overalloverall, on regularly refreshed questions#30of 31 · 43.9Best attop three among current models
Not in the top three among current models on any category we track.
Where to use itprices checked 27 Oct 2026
API providers 1 · per 1M tokens, in / out
Host prices from OpenRouter; going direct can differ.
Facts and sources
Released6 Oct 2026
Announcementmistral.ai ↗
Release entryMistral Large 4
Scored asMistral Large 4
Output speed116 tokens/s
Hugging Facenot linked
