AtlaAI
/

Selene-1-Mini-Llama-3.1-8B-GPTQ-W8A8

Text Generation

text-generation-inference

Inference Endpoints

8-bit precision

compressed-tensors

Model card Files Files and versions Community

Selene-1-Mini-Llama-3.1-8B-GPTQ-W8A8 / recipe.yaml

kaikaidai's picture

Upload recipe.yaml with huggingface_hub

cdfd9d6 verified 12 days ago

history blame contribute delete

200 Bytes

	DEFAULT_stage:
	DEFAULT_modifiers:
	SmoothQuantModifier: {smoothing_strength: 0.8}
	GPTQModifier:
	sequential_update: true
	scheme: W8A8
	targets: Linear
	ignore: [lm_head]