hf_quant_config.json · Codeflux/MiniCPM5-2B-FP8 at main

MiniCPM5-2B-FP8 / hf_quant_config.json

Upload folder using huggingface_hub

1bd9d94 verified 4 days ago

273 Bytes

{
    "producer": {
        "name": "modelopt",
        "version": "0.47.0"
    },
    "quantization": {
        "quant_algo": "FP8",
        "kv_cache_quant_algo": null,
        "exclude_modules": [
            "lm_head",
            "model.embed_tokens"
        ]
    }
}

	{
	"producer": {
	"name": "modelopt",
	"version": "0.47.0"
	},
	"quantization": {
	"quant_algo": "FP8",
	"kv_cache_quant_algo": null,
	"exclude_modules": [
	"lm_head",
	"model.embed_tokens"
	]
	}
	}