espressor
/

meta-llama.Llama-3.2-3B-Instruct_W8A8_FP8

Text Generation

text-generation-inference

Inference Endpoints

compressed-tensors

Model card Files Files and versions Community

meta-llama.Llama-3.2-3B-Instruct_W8A8_FP8 / recipe.yaml

xzyao's picture

Upload folder using huggingface_hub

ebc57d7 verified 30 days ago

history blame contribute delete

136 Bytes

	DEFAULT_stage:
	DEFAULT_modifiers:
	QuantizationModifier:
	ignore: [lm_head]
	targets: [Linear]
	scheme: FP8_DYNAMIC