Llama 3.2 Vision (90B)
Release Date: 4/18/2024
Benchmarked by 
Vision-enabled Llama model with 90B parameters for advanced use cases.
Avg. Accuracy
43.43%
Latency
30.70s
Cost (In/Out)
1.2 / 1.2
Context Window
131k
Max Output Tokens
4k
Input Modality
Hyperparameter settings
Default Provider :
N/A
Some benchmarks may use different provider and parameters. Please refer to the benchmark page for more information.
Temperature
Default
Top P
Default
Top K
Default
Max Output Tokens
4,096
Academic Benchmarks
Proprietary Benchmarks (contact us to get access)