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

Kernels:
webgpu-kernels
/
com.microsoft.GroupQueryAttention

Kernel card Files Files and versions
xet
Community
com.microsoft.GroupQueryAttention / build /webgpu
816 kB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 3 commits
Xenova's picture
Xenova HF Staff
sync 91d990483a17
78d4fcb verified 3 days ago
  • attention-rank4-tiled.wgsl.jinja
    5.43 kB
    sync 91d990483a17 3 days ago
  • attn-flash-decode-splitk-merge.wgsl.jinja
    5.5 kB
    sync 91d990483a17 3 days ago
  • attn-flash-decode-splitk.wgsl.jinja
    20.2 kB
    sync 91d990483a17 3 days ago
  • attn-flash-online.wgsl.jinja
    13.5 kB
    sync 91d990483a17 3 days ago
  • attn-flash-prefill-cluster.wgsl.jinja
    29.2 kB
    sync 91d990483a17 3 days ago
  • attn-flash-q32-broadcast.wgsl.jinja
    8.57 kB
    sync 91d990483a17 3 days ago
  • attn-materialized-rowstats-combine-f32.wgsl.jinja
    3.05 kB
    sync 91d990483a17 3 days ago
  • attn-materialized-sgmat-f32.wgsl.jinja
    23.9 kB
    sync 91d990483a17 3 days ago
  • attn-online-scalar.wgsl.jinja
    6.89 kB
    sync 91d990483a17 3 days ago
  • bench.json
    71.3 kB
    sync 91d990483a17 3 days ago
  • gqa-attention.wgsl.jinja
    11.7 kB
    sync 91d990483a17 3 days ago
  • gqa-present.wgsl.jinja
    19.4 kB
    sync 91d990483a17 3 days ago
  • gqa-qprep.wgsl.jinja
    2.54 kB
    sync 91d990483a17 3 days ago
  • manifest.json
    257 kB
    sync 91d990483a17 3 days ago
  • metadata.json
    9.6 kB
    sync 91d990483a17 3 days ago
  • test.json
    329 kB
    sync 91d990483a17 3 days ago