neuralmagic
/

Qwen2.5-VL-7B-Instruct-FP8-Dynamic

Image-Text-to-Text

text-generation-inference

compressed-tensors

Model card Files Files and versions Community

Qwen2.5-VL-7B-Instruct-FP8-Dynamic / recipe.yaml

mgoin's picture

Upload folder using huggingface_hub

26bec9d verified about 2 months ago

158 Bytes

	DEFAULT_stage:
	DEFAULT_modifiers:
	QuantizationModifier:
	ignore: ['re:.lm_head', 're:visual.']
	targets: [Linear]
	scheme: FP8_DYNAMIC