MiniMax-M2.7
MiniMax-M2.7 is an open-weights AI model from MiniMax with an index score of 132 (GPT-4 = 100), rank 121 of 217.
132114–151Index score#121 of 217 models
$0.21/ $0.84Price per million tokensinput / output · via OpenRouter
197KContext windowtokens it reads at once
18 Mar 2026ReleasedReplaced by MiniMax-M3
Where it sitson the frontier
o1 reached this score 15 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsno evidence yet
Codingevidence
Agentsno evidence yet
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 8 leaderboards
The Model Indexthird-party tests, our method#121of 217 · 132
LMArena · Codingcoding questions in chat#99of 357 · 1480LMArena · Mathsmaths questions in chat#105of 346 · 1423LMArena · Long promptslong, complex requests#106of 340 · 1431LMArena · Instruction followingfollowing detailed instructions#114of 362 · 1408LMArena · Textoverall chat quality, judged by people#115of 362 · 1415LMArena · Creative writingcreative writing#129of 360 · 1364LMArena · WebDevbuilding web apps and front ends#90of 133 · 1398Best attop three among current models
Replaced by MiniMax-M3; it no longer leads any category among current models.
Where to use itprices checked 18 Nov 2026
API providers 2 · per 1M tokens, in / out
Run it yourself
Official weightsMiniMaxAI/MiniMax-M2.7↗Host prices from OpenRouter; going direct can differ.
Facts and sources
Released18 Mar 2026
Announcementminimax.io ↗
Release entryMiniMax-M2.7
Scored asMiniMax-M2.7
Output speed–
Hugging FaceMiniMaxAI/MiniMax-M2.7 ↗