Loading...
Loading...
MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1 to extend into general office work, reaching fluency in generating and operating Word, Excel, and Powerpoint files, context switching between diverse software environments, and working across different agent and human teams. Scoring 80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench, and 76.3% on BrowseComp, M2.5 is also more token efficient than previous generations, having been trained to optimize its actions and output through planning.
Quality Index
41.9
34th of 442
Top 8%
Coding Index
37.4
37th of 352
Top 11%
Price/1M
$0.53
392nd cheapest
69% above median
Top 58%
Speed
48 tok/s
Top 49%
TTFT
3.87s
Context Window
197K
131st largest
Top 38%
Input
$0.30
per 1M tokens
Output
$1.20
per 1M tokens
Blended
$0.53
per 1M tokens
Cheaper than 42% of models. Median price is $0.31/1M tokens.
Daily
$0.53
Monthly
$15.75
48
tokens/sec
Faster than 51% of models
3.87
seconds
Faster than 10% of models
45.70
seconds
Faster than 4% of models
Market Median
46 tok/s
5% faster
Median TTFT
0.42s
827% slower
Throughput/Dollar
91
tok/s per $/1M
Speed Comparison
Context Window
197K
tokens
Larger than 62% of models
Max Output
66K
tokens
33% of context