neuralmagic
/

Llama-3.2-11B-Vision-Instruct-FP8-dynamic

Text Generation

compressed-tensors

Model card Files Files and versions Community

Llama-3.2-11B-Vision-Instruct-FP8-dynamic / recipe.yaml

mgoin's picture

Upload folder using huggingface_hub

c0ee264 verified about 2 months ago

192 Bytes

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