Any to Any Modality Model Training Framework. Lean for research and hacking.
active 2025-09-19 → 2026-07-22 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 307 days from 2025-09-19 to 2026-07-22. Pushes: 257 total, peak 18 in a day. Pull requests: 43 total, peak 3 in a day. Issues: 16 total, peak 2 in a day. Comments: 113 total, peak 12 in a day. Stars: 279 total, peak 61 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 |
|---|---|---|---|---|
| kcz358 | 208 | 109 | 23 | 38 |
| Luodian | 50 | 40 | 2 | 2 |
| pufanyi | 43 | 30 | 4 | 7 |
| Copilot | 37 | 0 | 0 | 25 |
| ngquangtrung57 | 33 | 19 | 2 | 9 |
| coderabbitai[bot] | 21 | 0 | 0 | 14 |
| KemingWu | 20 | 16 | 2 | 2 |
| chatgpt-codex-connector[bot] | 19 | 0 | 0 | 10 |
| mwxely | 13 | 4 | 3 | 4 |
| nssmd | 11 | 11 | 0 | 0 |
| github-actions[bot] | 10 | 10 | 0 | 0 |
| Jinghao-Guo | 10 | 8 | 2 | 0 |
| BIGKnight | 8 | 6 | 0 | 2 |
| JinjieNi | 4 | 4 | 0 | 0 |
| Purshow | 3 | 0 | 3 | 0 |
| wjn922 | 2 | 0 | 0 | 1 |
| oneScotch | 2 | 1 | 1 | 0 |
| Doctor-James | 2 | 0 | 0 | 0 |
| PercyHayes | 1 | 0 | 0 | 0 |
| liqiangniu | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Pull request#169kcz3582026-05-11 08:06
- Pull request#164kcz3582026-05-10 05:20
- Pull request#162kcz3582026-05-09 03:29
- Issue#128kcz3582026-04-27 06:17Some Issues when Training Qwen
- Pull request#159kcz3582026-04-27 06:17
- Pull request#159kcz3582026-04-27 06:16
- Issue#132kcz3582026-04-13 09:45qwen3vl的视频时间戳token位置与transformers库不符
- Pull request#154kcz3582026-04-13 09:45
- Pull request#153kcz3582026-04-10 14:53
- Issue comment#136kcz3582026-04-06 03:22About pre-training of diffusion language model
- Pull request#149Purshow2026-03-20 09:55
- Pull request#141Purshow2026-03-07 07:13
- Pull request#137Jinghao-Guo2026-02-25 15:08
- Issue#133shg82026-02-20 06:43Loss normalization in Bagel training
- Issue#135Luodian2026-02-16 07:28LMMs-Engine Upgrade Plan
- Issue#135kcz3582026-02-13 02:00LMMs-Engine Upgrade Plan
- Issue comment#133kcz3582026-02-09 05:44Loss normalization in Bagel training
- Issue comment#132kcz3582026-02-09 05:41qwen3vl的视频时间戳token位置与transformers库不符
- Pull request#134Purshow2026-02-09 04:54
- Pull request#129kcz3582026-01-14 08:43
- Pull request#129kcz3582026-01-14 08:43
- Pull request#127kcz3582026-01-12 02:01
- Issue#128pufanyi2026-01-09 17:15Same Issues when Training Qwen
- Issue#125Doctor-James2026-01-09 11:29About pretrain
- Issue comment#127kcz3582026-01-09 06:09[feat] Allow middle checkpoint evaluation in background using lmms-eval http server
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 279 stars here means stars gained during the window, not the repo's star count.