Private AI
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
Performance metrics and benchmarks
Sourced from LMArena.
Arena Score
1363.9
Overall Rank
#158 / 367
Votes
35,208
Confidence Interval
1359.6 - 1368.1
Coding
#152 / 362
6,486 votes
1416.4
Math
#153 / 356
1,801 votes
1372.1
Longer Query
#163 / 345
6,963 votes
1364.7
Creative Writing
#171 / 365
4,546 votes
1317.5
Instruction Following
#165 / 367
8,687 votes
1346.2
Hard Prompts
#157 / 367
15,831 votes
1381.1
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
French
#115 / 253
425 votes
1410.8
German
#132 / 275
828 votes
1359.9
Industry Mathematical
#145 / 350
1,786 votes
1383.1
Industry Medicine And Healthcare
#149 / 337
2,111 votes
1393.0
English
#150 / 367
16,873 votes
1385.2
Spanish
#153 / 254
716 votes
1353.5
Industry Life And Physical And Social Science
#154 / 365
5,883 votes
1383.9
Chinese
#156 / 337
1,821 votes
1385.5
Industry Software And It Services
#157 / 367
11,666 votes
1402.8
Exclude Ties
#159 / 367
24,819 votes
1329.5
Expert
#161 / 316
1,591 votes
1363.8
Multi Turn
#161 / 365
5,692 votes
1357.3
Hard Prompts English
#162 / 366
8,160 votes
1394.8
Industry Entertainment And Sports And Media
#162 / 365
6,296 votes
1325.1
Korean
#162 / 245
635 votes
1272.9
Russian
#165 / 330
1,929 votes
1348.9
Industry Business And Management And Financial Operations
#166 / 360
6,113 votes
1352.2
Industry Legal And Government
#172 / 340
2,408 votes
1364.9
Industry Writing And Literature And Language
#173 / 366
7,801 votes
1331.8
Japanese
#173 / 237
773 votes
1225.9
Published 2026-06-16 · Matched as minimax-m1
LMArena Dataset