mlx-community/Mistral-Small-24B-Instruct-2501-4bit

Quantization vs mistralai/Mistral-Small-24B-Instruct-2501. mlx-community/Mistral-Small-24B-Instruct-2501-4bit is a quantized version of the other (U32/F16).

5 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse