Mistral Medium 3.5
Mistral Medium 3.5 is an open-weights AI model from Mistral with an index score of 131 (GPT-4 = 100), rank 123 of 217.
131114–148Index score#123 of 217 models
$1.5/ $7.5Price per million tokensinput / output · via OpenRouter
262KContext windowtokens it reads at once
28 Apr 2026Released7 months ago
Where it sitson the frontier
o1 reached this score 17 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 10 leaderboards
The Model Indexthird-party tests, our method#123of 217 · 131
LMArena · Instruction followingfollowing detailed instructions#92of 362 · 1423LMArena · Creative writingcreative writing#93of 360 · 1399LMArena · Textoverall chat quality, judged by people#99of 362 · 1426LMArena · Codingcoding questions in chat#100of 357 · 1478LMArena · Mathsmaths questions in chat#97of 346 · 1426LMArena · Long promptslong, complex requests#109of 340 · 1429Epoch Capabilities Indexoverall capability from many benchmarks#112of 231 · 141.6LMArena · Visionunderstanding images#80of 145 · 1199LMArena · WebDevbuilding web apps and front ends#117of 133 · 1263Best attop three among current models
Not in the top three among current models on any category we track.
Where to use itprices checked 18 Nov 2026
API providers 1 · per 1M tokens, in / out
Run it yourself
Official weightsmistralai/Mistral-Medium-3.5-128B↗Host prices from OpenRouter; going direct can differ.
Facts and sources
Released28 Apr 2026
Announcementmistral.ai ↗
Release entryMistral Medium 3.5
Scored asMistral Medium 3.5
Output speed–
Hugging Facemistralai/Mistral-Medium-3.5-128B ↗