Llama 3.3 70B Instruct FP8 Fast
model-card · v1.0.0
Llama 3.3 70B quantized to fp8 precision and optimized for faster inference.
Raw atom
/atoms/model-card/cf-meta-llama-3.3-70b-instruct-fp8-fast.json · schema
model-card · v1.0.0
Llama 3.3 70B quantized to fp8 precision and optimized for faster inference.
/atoms/model-card/cf-meta-llama-3.3-70b-instruct-fp8-fast.json · schema