neuralmagic
/

Qwen2.5-72B-quantized.w8a8

Text Generation

8-bit precision

compressed-tensors

Model card Files Files and versions Community

Qwen2.5-72B-quantized.w8a8 / recipe.yaml

alexmarques's picture

Upload folder using huggingface_hub

c471080 verified 5 months ago

history blame contribute delete

191 Bytes

	quant_stage:
	quant_modifiers:
	GPTQModifier:
	sequential_update: true
	dampening_frac: 0.1
	ignore: [lm_head]
	scheme: W8A8
	targets: Linear
	observer: mse