Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

nvidia
/
Qwen3.8-Flash-Next-FP4-Local

Image-Text-to-Text
Safetensors
Model Optimizer
qwen4_exp
nvidia
ModelOpt
Qwen3.8
quantized
FP4
fp4
conversational
8-bit precision
modelopt
Model card Files Files and versions
xet
Community
16
New discussion
Resources
  • PR & discussions documentation
  • Code of Conduct
  • Hub documentation

Runs on stock vLLM v0.34.0 (no b12x) on one DGX Spark β€” recipe and one gotcha

#16 opened about 22 hours ago by
CTala

Will you make the same job for Qwen4.0 Flash?

#15 opened 4 days ago by
moranilt

Byte indentical to local-inference-lab model

#14 opened 7 days ago by
vvekthkr

DGX Spark: load OOM fixed with PYTORCH_ALLOC_CONF, plus results at 131K context

#13 opened 14 days ago by
howellx

Running FP4-Local on 4 Γ— V100 32 GB (Volta) with ReVolta

#12 opened 24 days ago by
leoncca

Recipe for 4x RTX Pro 4000

#11 opened 24 days ago by
jmig1109

Serving FP4-Local on SGLang (one RTX PRO 6000, PRs open for review)

πŸ‘ 1
1
#10 opened 27 days ago by
AndreasBis

Please implement Thor Devkit deployment

πŸ‘ 2
#9 opened 27 days ago by
the-z0rb

made my DGX Spark more valuable

#7 opened about 1 month ago by
gmrxde

Merci !

#6 opened about 1 month ago by
QuentinReytinas

Docker (WSL2)?

5
#5 opened about 1 month ago by
evilperson068

Recipe for VLLM dev (no docker container), and Humaneval benchmark results vs RadixArk

πŸ‘ 11
2
#4 opened about 1 month ago by
OverleapSecreted
Company
TOS Privacy About Careers
Website
Models Datasets Spaces Pricing Docs