FAIR Sequence Modeling Toolkit 2,fairseq2 is a sequence modeling toolkit that allows researchers and developers to train custom models for translation, summarization, language modeling, and other content generation tasks.
active 2023-08-22 → 2026-07-06 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 1050 days from 2023-08-22 to 2026-07-06. Pushes: 2,984 total, peak 56 in a day. Pull requests: 2,123 total, peak 28 in a day. Issues: 264 total, peak 20 in a day. Comments: 1,517 total, peak 48 in a day. Stars: 1,002 total, peak 109 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 |
|---|---|---|---|---|
| cbalioglu | 4.3K | 1.6K | 1.7K | 541 |
| artemru | 517 | 145 | 26 | 169 |
| jacklanchantin | 394 | 249 | 33 | 55 |
| zyaoj | 238 | 103 | 44 | 39 |
| uralik | 224 | 104 | 45 | 27 |
| kauterry | 211 | 81 | 29 | 55 |
| shagunsodhani | 176 | 3 | 3 | 77 |
| github-actions[bot] | 174 | 174 | 0 | 0 |
| antoine-tran | 171 | 97 | 13 | 42 |
| cirquit | 167 | 43 | 7 | 62 |
| gilkeren1 | 113 | 70 | 22 | 10 |
| MartinGleize | 93 | 15 | 14 | 32 |
| mattsetz | 88 | 35 | 11 | 20 |
| Ahmedsaed | 87 | 0 | 12 | 43 |
| rsyue | 78 | 56 | 4 | 8 |
| asn1814 | 77 | 45 | 15 | 9 |
| gwenzek | 77 | 16 | 2 | 31 |
| swarnaHub | 46 | 29 | 2 | 6 |
| meta-cla[bot] | 39 | 0 | 38 | 1 |
| YunchaoYang | 37 | 33 | 0 | 3 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#1506meta-cla[bot]2026-04-04 20:32Fix typo in WerMetric: use hyp_seqs instead of ref_seqs for hyp_seqs_list
- Pull request#1495cirquit2026-03-25 21:23
- Issue comment#1496cirquit2026-03-25 19:08Add Gemma3n model family to fairseq2
- Issue comment#1126cirquit2026-03-16 21:44adding integration with fsspec FileSystem
- Pull request#1495cirquit2026-03-14 02:47
- Issue#1494veselinve2026-03-08 22:24Add the Windows native support i want to try the omnilingual-asr
- Issue comment#1479rsyue2026-02-17 23:19HG - Hardware-specific sharding test and instructions with torchrun
- Pull request#1490artemru2026-02-06 12:46
- Pull request#1491cbalioglu2026-02-05 21:12
- Pull request#1490artemru2026-02-04 12:16
- Pull request#1488lydiadli2026-02-02 03:48
- Pull request#1488meta-cla[bot]2026-02-02 03:41
- Issue comment#1126artemru2026-01-26 23:23adding integration with fsspec FileSystem
- Issue comment#1126artemru2026-01-26 19:35adding integration with fsspec FileSystem
- Issue#1486cirquit2026-01-16 22:15Iterating through asset cards fails when unrelated asset is broken
- Issue#1486cirquit2026-01-16 22:15Iterating through asset cards fails when unrelated asset is broken
- Pull request#1484meta-cla[bot]2026-01-09 18:49
- Issue#1483rsyue2026-01-06 22:06Improve fbpkg integration
- Issue#1483rsyue2026-01-06 22:06Improve fbpkg integration
- Issue comment#1481cirquit2026-01-02 16:19Different batch sizes result in large numerical differences in transformer models
- Issue comment#1481gilkeren12025-12-30 23:49Different batch sizes result in large numerical differences in transformer models
- Issue#1481cirquit2025-12-30 23:18Different batch sizes result in large numerical differences in transformer models
- Issue comment#1481gilkeren12025-12-30 00:45Different batch sizes result in large numerical differences in transformer models
- Issue#1481gilkeren12025-12-29 23:49Different batch sizes result in large numerical differences in transformer models
- Issue#1481gilkeren12025-12-29 23:49Different batch sizes result in large numerical differences in transformer models
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 1,002 stars here means stars gained during the window, not the repo's star count.