HuggingFaceTB/SmolLM2-1.7B-Instruct-Q8-mlx

Quantization vs HuggingFaceTB/SmolLM2-1.7B-Instruct. HuggingFaceTB/SmolLM2-1.7B-Instruct-Q8-mlx is a quantized version of the other (U32/F16).

On the Hub · Downloads on Model Pulse