首个端到端的视频扩散 Transformer，它能够在参考图像和音频的条件下，合成无限长度的高质量音频驱动虚拟形象视频，且无需任何后处理 We present StableAvatar, the first end-to-end video diffusion transformer, which synthesizes infinite-length high-quality audio-driven avatar videos without any post-processing, conditioned on a reference image and audio.
active 2025-08-12 → 2026-04-27 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 259 days from 2025-08-12 to 2026-04-27. Pushes: 21 total, peak 6 in a day. Pull requests: 12 total, peak 2 in a day. Issues: 87 total, peak 6 in a day. Comments: 120 total, peak 13 in a day. Stars: 627 total, peak 68 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 |
|---|---|---|---|---|
| Francis-Rings | 123 | 20 | 6 | 61 |
| zhangquanwei962 | 10 | 0 | 0 | 9 |
| yqxd | 10 | 0 | 0 | 5 |
| QUTGXX | 7 | 0 | 0 | 6 |
| YinmingHuang | 6 | 1 | 2 | 3 |
| XuJianzhi | 6 | 0 | 0 | 1 |
| blaji-villeb106 | 6 | 0 | 0 | 3 |
| Eetenal | 5 | 0 | 0 | 4 |
| ghx2757 | 5 | 0 | 0 | 2 |
| GestureDiffuMamba | 4 | 0 | 0 | 3 |
| liwang0621 | 4 | 0 | 0 | 0 |
| gluttony-10 | 3 | 0 | 2 | 1 |
| smthemex | 3 | 0 | 0 | 2 |
| cosmicrealm | 2 | 0 | 0 | 2 |
| arshia-dh | 2 | 0 | 0 | 1 |
| nitinmukesh | 2 | 0 | 0 | 2 |
| jylovec | 2 | 0 | 0 | 2 |
| kasf666 | 2 | 0 | 0 | 1 |
| zhecks | 2 | 0 | 0 | 2 |
| Dazzastrous | 2 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#90Francis-Rings2026-01-20 01:45训练时video的fps和audio的重采样率
- Issue#89Francis-Rings2026-01-16 02:18Result error
- Issue comment#89Francis-Rings2026-01-16 02:17Result error
- Issue#89sebastianopazo12026-01-15 18:52Result error
- Issue#88Francis-Rings2025-12-25 01:54hello，这个能支持「视频+音频」生成对嘴视频吗
- Issue comment#88Francis-Rings2025-12-25 01:54hello，这个能支持「视频+音频」生成对嘴视频吗
- Issue#87chengxumiaodaren2025-12-23 15:18中文效果明显不如英文
- Issue#86Francis-Rings2025-12-18 06:37Sync-C↑ Sync-D↓指标
- Issue#86wokaodeshang2025-12-17 09:02Sync-C↑ Sync-D↓指标
- Issue#85Francis-Rings2025-12-09 08:51Confusion about FPS
- Issue#85zhenye2342025-12-08 06:27Confusion about FPS
- Issue#84Francis-Rings2025-12-07 01:41哥 论文里的训练总共要花多少步呢 还有1.3b的模型训练的时候大概多少秒一步呢
- Issue comment#84Francis-Rings2025-12-05 01:51哥 论文里的训练总共要花多少步呢 还有1.3b的模型训练的时候大概多少秒一步呢
- Issue#84zhenye2342025-12-04 09:02哥 论文里的训练总共要花多少步呢 还有1.3b的模型训练的时候大概多少秒一步呢
- Issue comment#83YinmingHuang2025-12-04 05:37CUDA out of memory. Tried to allocate 62.02 GiB
- Issue#83fallbernana1234562025-12-03 03:28CUDA out of memory. Tried to allocate 62.02 GiB
- Issue#82weirui04302025-11-19 06:25输入音频是中文音频，输出结果出错
- Issue comment#59blaji-villeb1062025-11-15 16:56Add multi-language interface support and gitignore
- Issue#81blaji-villeb1062025-11-15 16:55请澄清项目 Star 数激增问题
- Issue#80blaji-villeb1062025-11-15 16:55该项目疑似刷星，请官方给个说法
- Issue#79blaji-villeb1062025-11-15 16:54星数疑似异常，恳请作者给个说明
- Issue comment#36blaji-villeb1062025-11-15 16:4814B模型LoRA训练报错
- Issue comment#77blaji-villeb1062025-11-15 16:47قهوه
- Issue#77mohammadazi148-cell2025-11-14 17:18قهوه
- Issue#76jshq19712025-11-14 09:36erica
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 627 stars here means stars gained during the window, not the repo's star count.