dgr237
/

Qwopus3.6-35B-A3B-Coder-NVFP4

Mixture of Experts

8-bit precision

compressed-tensors

Model card Files Files and versions

Qwopus3.6-35B-A3B-Coder-NVFP4 / recipe.yaml

dgr237's picture

Upload recipe.yaml with huggingface_hub

242d1e2 verified 6 days ago

206 Bytes

	default_stage:
	default_modifiers:
	QuantizationModifier:
	targets: [Linear]
	ignore: [lm_head, 're:.ssm.', 're:.linear_attn.']
	scheme: NVFP4
	bypass_divisibility_checks: false