PyTorch native quantization and sparsity for training and inference
active 2025-03-16 → 2026-08-10 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 513 days from 2025-03-16 to 2026-08-10. Pushes: 5,407 total, peak 132 in a day. Pull requests: 2,305 total, peak 29 in a day. Issues: 298 total, peak 13 in a day. Comments: 5,143 total, peak 84 in a day. Stars: 422 total, peak 8 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
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| jerryzh168 | 3.2K | 728 | 236 | 1.1K |
| vkuzo | 1.9K | 985 | 187 | 397 |
| danielvegamyhre | 1.8K | 950 | 393 | 292 |
| pytorch-bot[bot] | 940 | 0 | 107 | 826 |
| andrewor14 | 837 | 399 | 77 | 190 |
| github-actions[bot] | 737 | 734 | 0 | 3 |
| drisspg | 632 | 170 | 51 | 203 |
| jcaip | 456 | 261 | 71 | 61 |
| metascroy | 451 | 123 | 82 | 120 |
| Xia-Weiwen | 422 | 30 | 47 | 225 |
| meta-cla[bot] | 366 | 0 | 319 | 47 |
| facebook-github-bot | 340 | 41 | 43 | 256 |
| jainapurva | 333 | 171 | 65 | 47 |
| howardzhang-cv | 314 | 245 | 34 | 22 |
| namgyu-youn | 271 | 0 | 24 | 182 |
| liangan1 | 209 | 15 | 62 | 78 |
| liangel-02 | 177 | 106 | 38 | 13 |
| kimishpatel | 119 | 40 | 5 | 48 |
| lisjin | 118 | 68 | 16 | 17 |
| Valentine233 | 110 | 11 | 16 | 51 |
Recent activity
Latest issues, pull requests and releases
- Pull request#4722vkuzo2026-08-07 19:50
- Pull request#4717vkuzo2026-08-07 19:40
- Issue comment#4726vkuzo2026-08-07 19:04Delete dead prototype.common module and broken benchmark scripts
- Issue comment#4725pytorch-bot[bot]2026-08-07 18:39Refine MXFP8 CUDA quantization semantics
- Pull request#4723vkuzo2026-08-07 18:29
- Pull request#4722vkuzo2026-08-07 18:29
- Issue comment#4722vkuzo2026-08-07 18:15Remove model-release CI workflow and its scripts
- Pull request#4722vkuzo2026-08-07 18:15
- Issue comment#4718claude[bot]2026-08-07 17:52Move int8 matmul kernels into the int8 workflow and drop them from public API
- Issue comment#4709vkuzo2026-08-07 17:44Enable rowwise MXFP8 CUDA quantization
- Pull request#4721vkuzo2026-08-07 17:33
- Pull request#4720vkuzo2026-08-07 17:29
- Pull request#4721meta-cla[bot]2026-08-07 17:29
- Issue comment#4720pytorch-bot[bot]2026-08-07 17:29Delete torchao/_models (SAM/SAM2) and its orphaned consumers
- Issue comment#4721vkuzo2026-08-07 17:29Remove dead consumers of deleted torchao/_models.llama and _eval
- Pull request#4720meta-cla[bot]2026-08-07 17:29
- Issue comment#4720vkuzo2026-08-07 17:29Delete torchao/_models (SAM/SAM2) and its orphaned consumers
- Pull request#4720vkuzo2026-08-07 17:29
- Pull request#4719vkuzo2026-08-07 16:19
- Pull request#4719meta-cla[bot]2026-08-07 16:18
- Pull request#4718meta-cla[bot]2026-08-07 16:18
- Issue comment#4718pytorch-bot[bot]2026-08-07 16:18Move int8 matmul kernels into the int8 workflow and drop them from public API
- Issue comment#4719vkuzo2026-08-07 16:18Split blockwise fp8 kernels out of torchao/kernel into their consumers
- Issue comment#4718vkuzo2026-08-07 16:18Move int8 matmul kernels into the int8 workflow and drop them from public API
- Pull request#4717vkuzo2026-08-07 16:18
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 422 stars here means stars gained during the window, not the repo's star count.