novita
/

Deepseek-R1-W4AFP8

8-bit precision

Model card Files Files and versions

Deepseek-R1-W4AFP8 / hf_quant_config.json

Cornmonster's picture

Upload folder using huggingface_hub

1f608e0 verified 7 months ago

108 Bytes

	{
	"quantization": {
	"quant_algo": "MIXED_PRECISION",
	"kv_cache_quant_algo": null
	}
	}