Accuracy (Average)
37.44%
Latency (Average)
2.90s
Avg. Cost (In/Out)
0.2 / 0.4
Context Window
256k
Max Output Tokens
4k
Input Modality
Hyperparameter settings
Default Provider :
AI21 Labs
Some benchmarks may use different provider and parameters. Please refer to the benchmark page for more information.
Temperature
0.4
Top P
Default
Top K
Default
Max Output Tokens
4,096
Show rankings only among open weight models
Benchmarks
Accuracy
Rankings
Contact us
Proprietary Benchmarks (contact us to get access)
Academic Benchmarks
Read about our methodology.