Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • 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

Spaces:
webml-community
/
gemma-4-webgpu-kernels
Running

App Files Files Community
13
New discussion
Resources
  • PR & discussions documentation
  • Code of Conduct
  • Hub documentation

Fix NVIDIA/Windows output corruption: add the missing sgExact32 guard to 8 bare subgroupAdd sites

❤️ 1
#13 opened 2 days ago by
Ar5en1c

model.generate kwargs

1
#12 opened 10 days ago by
pythiccoder

Model generating gibberish: THIS IS A STRONG SIGNAL!

#11 opened 12 days ago by
DeathDaDev

Model Panic.

👍 2
1
#10 opened 14 days ago by
qsardor

gema

#9 opened 14 days ago by
NNNNautiluss

Fix Windows WebGPU subgroup generation corruption

3
#8 opened 17 days ago by
igorls

Windows_Chrome_Fix

#7 opened 19 days ago by
specimba

Hallucination

3
#6 opened 19 days ago by
johnlockejrr

Iphone 17 Pro Support?

#5 opened 23 days ago by
pythiccoder

Licensing / reuse of your gemma-4-e2b.js (Gemma4Mobile) WebGPU runtime

1
#4 opened 24 days ago by
AlbertJW

are you it is working ?

8
#3 opened 25 days ago by
specimba

DecodeOprojNorm variant rejected on NVIDIA Ampere: subgroupMaxSize is 128

16
#1 opened 26 days ago by
Intellipedia
Company
TOS Privacy About Careers
Website
Models Datasets Spaces Pricing Docs