Mistral Small 3.2
Mistral Small 3.2 is an open-weights AI model from Mistral with an index score of 117 (GPT-4 = 100), rank 161 of 217.
117101–134Index score#161 of 217 models
$0.075/ $0.20Price per million tokensinput / output · via OpenRouter
256KContext windowtokens it reads at once
20 Jun 2025ReleasedReplaced by Mistral Small 4
Where it sitson the frontier
o1-mini reached this score 9 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingno evidence yet
Agentsevidence
Professional workno evidence yet
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 8 leaderboards
The Model Indexthird-party tests, our method#161of 217 · 117
LMArena · Codingcoding questions in chat#164of 357 · 1412LMArena · Creative writingcreative writing#170of 360 · 1323LMArena · Textoverall chat quality, judged by people#171of 362 · 1357LMArena · Instruction followingfollowing detailed instructions#180of 362 · 1337LMArena · Long promptslong, complex requests#174of 340 · 1358LMArena · Mathsmaths questions in chat#181of 346 · 1338LMArena · Visionunderstanding images#106of 145 · 1140Best attop three among current models
Replaced by Mistral Small 4; it no longer leads any category among current models.
Where to use itprices checked 18 Nov 2026
API providers 3 · per 1M tokens, in / out
Host prices from OpenRouter; going direct can differ.
Facts and sources
Released20 Jun 2025
Announcementnot linked yet
Release entry–
Scored asMistral Small 3.2
Output speed–
Hugging Facemistralai/Mistral-Small-3.2-24B-Instruct-2506 ↗