mlx-community/Llama-3.1-Nemotron-8B-UltraLong-1M-Instruct-6bit

Quantization vs nvidia/Llama-3.1-Nemotron-8B-UltraLong-1M-Instruct. mlx-community/Llama-3.1-Nemotron-8B-UltraLong-1M-Instruct-6bit is a quantized version of the other (U32/F32).

On the Hub · Downloads on Model Pulse