MiniMax-M1 is a hybrid MoE reasoning model with 40K thinking budget. World's first open-weight, large-scale hybrid-attention model with lightning attention for efficient test-time compute scaling. Excels at complex tasks requiring extensive reasoning.
Added Jan 8, 2025
Context Window
1.0M
Max Output
131.1K
Input Price (Auto)
$0.14/1M
Output Price (Auto)
$1.33/1M
Cache Read (Auto)
$0.070/1M
Benchmarks
Benchmarks
Performance metrics and benchmarks
Sourced from LMArena.
Arena Score
1363.7
Overall Rank
#186 / 395
Votes
34,014
Confidence Interval
1359.4 - 1368.0
Category Scores
Coding
#179 / 390
6,332 votes
1416.4
Math
#177 / 379
1,762 votes
1369.2
Longer Query
#192 / 373
6,899 votes
1362.9
Creative Writing
#196 / 393
4,452 votes
1319.0
Instruction Following
#191 / 395
8,541 votes
1345.9
Hard Prompts
#185 / 395
15,484 votes
1380.4
Additional Categories22
French
#152 / 276
465 votes
1394.9
German
#152 / 296
877 votes
1362.6
Polish
#163 / 219
3,518 votes
1352.6
Spanish
#168 / 277
824 votes
1366.8
Industry Mathematical
#172 / 374
1,733 votes
1378.2
Korean
#174 / 265
659 votes
1282.9
Industry Medicine And Healthcare
#176 / 363
2,081 votes
1391.1
English
#179 / 395
14,821 votes
1384.7
Chinese
#182 / 367
2,345 votes
1386.4
Industry Life And Physical And Social Science
#185 / 393
5,747 votes
1381.6
Industry Software And It Services
#185 / 395
11,112 votes
1402.5
Exclude Ties
#186 / 395
23,907 votes
1328.8
Industry Entertainment And Sports And Media
#186 / 393
6,176 votes
1325.2
Hard Prompts English
#187 / 393
6,942 votes
1395.4
Multi Turn
#187 / 393
5,422 votes
1357.1
Russian
#188 / 359
2,349 votes
1350.8
Expert
#192 / 345
1,572 votes
1362.4
Japanese
#192 / 261
825 votes
1232.9
Non English
#192 / 395
19,184 votes
1340.1
Industry Business And Management And Financial Operations
#195 / 388
5,984 votes
1351.6
Industry Writing And Literature And Language
#197 / 394
7,636 votes
1331.9
Industry Legal And Government
#200 / 368
2,330 votes
1363.8
Published 2026-08-27 · Matched as minimax-m1
LMArena DatasetProviders
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
Related text models
Compare MiniMax M1 with similar models from the same provider or model family.
MiniMax M3
minimax/minimax-m3MiniMax M3 is the non-thinking route for MiniMax's open-weights frontier model, built for coding, agent workflows, tool use, and multimodal understanding from step zero. It keeps native thinking disabled for faster direct answers. MiniMax reports 59.0% on SWE-Bench Pro and 66.0% on Terminal Bench 2.1, with Sparse Attention designed to scale context to 1M. It starts with a 512K context cap on NanoGPT for now.
MiniMax M3 Thinking
minimax/minimax-m3:thinkingMiniMax M3 Thinking is the adaptive-thinking version of MiniMax's open-weights frontier model for coding, agent workflows, tool use, long-context tasks, and native multimodal understanding. MiniMax reports 59.0% on SWE-Bench Pro and 66.0% on Terminal Bench 2.1, with Sparse Attention designed to scale context to 1M. It starts with a 512K context cap on NanoGPT for now.
MiniMax Latest
minimax/minimax-latestCompatibility alias that routes to the newest MiniMax text model. Currently routes to MiniMax M3 (adaptive thinking).
MiniMax M2.7
minimax/minimax-m2.7MiniMax M2.7 is the first model deeply involved in iterating on its own training. It excels in real-world software engineering (SWE-Pro 56.22%), end-to-end project delivery (VIBE-Pro 55.6%), and complex office workflows with strong tool-use compliance and agentic capabilities.
MiniMax M2.7 Turbo
minimax/minimax-m2.7-turboMiniMax M2.7 Turbo is the highspeed and higher priced route for M2.7.
MiniMax M2.5
minimax/minimax-m2.5MiniMax M2.5 is a productivity-focused flagship model that builds on M2.1 with stronger coding and real-world office workflow performance (Word, Excel, PowerPoint), plus better tool-use planning and token efficiency.