RedHatAI/Meta-Llama-3.1-405B-Instruct-FP8

Quantization vs meta-llama/Llama-3.1-405B-Instruct. RedHatAI/Meta-Llama-3.1-405B-Instruct-FP8 is a quantized version of the other (F8_E4M3/BF16).

On the Hub · Downloads on Model Pulse