mlx-community/granite-4.0-h-350m-8bit
Quantization vs ibm-granite/granite-4.0-h-350m. mlx-community/granite-4.0-h-350m-8bit is a quantized version of the other (U32/BF16).
Quantization vs ibm-granite/granite-4.0-h-350m. mlx-community/granite-4.0-h-350m-8bit is a quantized version of the other (U32/BF16).