PIA-SPACE-LAB
/

Qwen3-VL-Embedding-2B-FP8

Feature Extraction

image-text-to-text

multimodal embedding

compressed-tensors

Model card Files Files and versions

Qwen3-VL-Embedding-2B-FP8 / recipe.yaml

jordaniel29's picture

Upload 14 files

d0f0def verified about 1 month ago

history blame contribute delete

160 Bytes

	default_stage:
	default_modifiers:
	QuantizationModifier:
	targets: [Linear]
	ignore: [lm_head, 're:model\.visual\..*']
	scheme: FP8_DYNAMIC