You can check the available precision types for any model using firectl. Models may support different numerical precisions like FP16, FP8, BF16, or INT8, which affect memory usage and inference speed.

​
Checking Default Precision

# Check default precision for a specific model
firectl get model accounts/fireworks/models/llama-v3p1-8b-instruct | grep "Default Precision"

​
Checking Supported Precisions

# Check supported precisions for a model
firectl get model accounts/fireworks/models/llama-v3p1-8b-instruct | grep -E "(Supported Precisions|Supported Precisions With Calibration)"