Qwen3.5-397B-A17B-NVFP4-AutoRound-4L

A concise reduced-layer NVFP4 checkpoint in the Reduced Models set.

Overview

  • Variant: NVFP4
  • Layer setting: 4L
  • Runtime target: Transformers

Origin

Validation

  • Inference validation: verified on Transformers inference flow.

Notes

  • Generated text may look garbled for some prompts in current reduced variants.
  • This output-quality issue is known and can be ignored for now.
Downloads last month
108
Safetensors
Model size
28B params
Tensor type
BF16
ยท
U8
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound-4L

Quantized
(87)
this model

Collection including intel-ai/Qwen3.5-397B-A17B-NVFP4-AutoRound-4L