nm-testing
/

Mistral-Small-3.1-24B-Instruct-2503-W4A16-G128

compressed-tensors

Model card Files Files and versions

Mistral-Small-3.1-24B-Instruct-2503-W4A16-G128 / recipe.yaml

kylesayrs's picture

Upload folder using huggingface_hub

304a295 verified 21 days ago

history blame contribute delete

228 Bytes

	default_stage:
	default_modifiers:
	GPTQModifier:
	targets: [Linear]
	ignore: ['re:.lm_head', 're:vision_tower.', 're:multi_modal_projector.*']
	scheme: W4A16
	sequential_targets: [MistralDecoderLayer]