Accuracy (Vals Index)
50.17% ± 1.50
Latency (Vals Index)
574.43s
Cost/Test (Vals Index)
$0.19
Context Window
262k
Max Output Tokens
262k
Input Modality
Hyperparameter settings
Default Provider :
Thinkingmachines
Some benchmarks may use different provider and parameters. Please refer to the benchmark page for more information.
Temperature
1
Top P
1
Top K
Default
Max Output Tokens
262,000
Reasoning Effort
0.99
Show rankings only among open weight models
Benchmarks
Accuracy
Rankings
Contact us
Read our methodology.