Skip to content

On-device AI across mobile, embedded and edge for PyTorch

active 2025-01-042026-08-11 (UTC)

Partial coverage12,291 / 13,997 hourly files (88%) · 2 absent upstream · 1,701 failed, retryable2025-01-042026-08-11 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
77.4K
Pushes
22.7K
Pull requests
9.5K
Issues
1.2K
Stars
964
Forks
221

Activity over time

Daily event counts in the loaded window

Line chart, 585 days from 2025-01-04 to 2026-08-11. Pushes: 22,720 total, peak 315 in a day. Pull requests: 9,472 total, peak 90 in a day. Issues: 1,222 total, peak 24 in a day. Comments: 22,568 total, peak 220 in a day. Stars: 964 total, peak 70 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

ContributorContributionsPushesPRsComments
facebook-github-bot6.1K1.1K8594.1K
github-actions[bot]5.9K2.1K223.8K
pytorch-bot[bot]4.2K01634K
GregoryComer3.5K2.4K319361
swolchok2.7K1.6K360590
SS-JIA2K1.4K286122
Gasoonjia1.8K1.2K200188
meta-codesync[bot]1.7K649687314
lucylq1.6K884273203
digantdesai1.6K24387619
larryliu08201.4K507196265
cccclai1.4K392188483
mergennachin1.3K487179341
kirklandsign1.3K554337171
Copilot1.3K4037725
kimishpatel1.3K63777327
pytorchbot1.2K228483500
shoumikhin1.1K597268134
meta-cla[bot]1.1K01K60
manuelcandales1.1K69710263

Recent activity

Latest issues, pull requests and releases

  • Issue comment#21677meta-cla[bot]2026-08-08 01:58
    [ExecuTorch][WebGPU] Add Gemma 4 speculative decode runtime
  • Issue comment#21679meta-cla[bot]2026-08-08 01:49
    [ExecuTorch][WebGPU] Optimize Qwen3 long-context FlashDecoding
  • Issue comment#21672github-actions[bot]2026-08-07 18:57
    [ExecuTorch][WebGPU] Add Gemma 4 plain runtime and guarded routes
  • Issue comment#21677pytorch-bot[bot]2026-08-07 18:57
    [ExecuTorch][WebGPU] Add Gemma 4 speculative decode runtime
  • Issue comment#21676pytorch-bot[bot]2026-08-07 18:57
    [ExecuTorch][WebGPU] Add Gemma 4 MTP export path
  • Issue comment#21675pytorch-bot[bot]2026-08-07 18:57
    [ExecuTorch][WebGPU] Add Gemma 4 MTP operator and route support
  • Pull request#21673JCNTH2026-08-07 18:56
  • Issue comment#21672pytorch-bot[bot]2026-08-07 18:56
    [ExecuTorch][WebGPU] Add Gemma 4 plain runtime and guarded routes
  • Pull request#21671JCNTH2026-08-07 18:56
  • Issue comment#21670pytorch-bot[bot]2026-08-07 18:56
    [ExecuTorch][WebGPU] Add shared model runtime prerequisites
  • Pull request#21670meta-cla[bot]2026-08-07 18:56
  • Pull request#21646pytorchbot2026-08-07 17:58
  • Issue comment#21668pytorch-bot[bot]2026-08-07 17:55
    Build and publish CUDA wheels
  • Pull request#21651pytorchbot2026-08-07 17:50
  • Pull request#21652pytorchbot2026-08-07 17:49
  • Pull request#21653pytorchbot2026-08-07 17:49
  • Pull request#21656pytorchbot2026-08-07 17:49
  • Pull request#21666pytorchbot2026-08-07 17:32
  • Pull request#21667meta-cla[bot]2026-08-07 17:24
  • Pull request#21667pytorchbot2026-08-07 17:24
  • Issue comment#21667pytorch-bot[bot]2026-08-07 17:23
    [ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction
  • Pull request#21667pytorchbot2026-08-07 17:23
  • Issue comment#21665pytorch-bot[bot]2026-08-07 17:23
    [ExecuTorch][WebGPU] Support general groups=1 Conv1d
  • Pull request#21665meta-cla[bot]2026-08-07 17:23
  • Pull request#21665pytorchbot2026-08-07 17:23

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 964 stars here means stars gained during the window, not the repo's star count.