Helios: Real Real-Time Long Video Generation Model
Python · active 2026-03-04 → 2026-06-27 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 116 days from 2026-03-04 to 2026-06-27. Pushes: 46 total, peak 8 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 21 total, peak 4 in a day. Comments: 65 total, peak 5 in a day. Stars: 388 total, peak 48 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 |
|---|---|---|---|---|
| SHYuanBest | 75 | 44 | 0 | 29 |
| splinter21 | 10 | 0 | 0 | 7 |
| Iriya99 | 4 | 0 | 0 | 3 |
| qqingzheng | 4 | 2 | 0 | 1 |
| charlotte1988q50 | 3 | 0 | 0 | 3 |
| nanwanj | 3 | 0 | 0 | 2 |
| skywalker00001 | 3 | 0 | 0 | 3 |
| OrangePure | 3 | 0 | 0 | 2 |
| stewart-jackson813 | 2 | 0 | 0 | 2 |
| 963658029 | 2 | 0 | 0 | 2 |
| Jumponthemoon | 2 | 0 | 0 | 1 |
| IceIce1ce | 2 | 0 | 0 | 2 |
| xiaomafei | 2 | 0 | 0 | 0 |
| ChirsHuanhuan | 2 | 0 | 0 | 1 |
| batman007herewego | 1 | 0 | 0 | 0 |
| liwang0621 | 1 | 0 | 0 | 1 |
| zhw-zhang | 1 | 0 | 0 | 0 |
| realhouse226 | 1 | 0 | 0 | 0 |
| shunshun-521 | 1 | 0 | 0 | 0 |
| yoonkhsc | 1 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#127SHYuanBest2026-06-09 03:02关于stage3 post阶段训练的显存占用
- Issue comment#126SHYuanBest2026-06-07 09:39Incorrect frame-aware history corruption ?
- Issue comment#115Xiaojiu-z2026-06-04 08:23逐chunk推理跳变的问题
- Issue comment#118bmakery2026-05-31 15:13是否可以跳过stage2，直接进行stage3？
- Issue#114striveAgain2026-05-25 02:48About the pretrained_model_name_or_path in Stage 1 init.
- Issue comment#1139636580292026-05-17 05:44B200/B300 support: current FA4 attention backend underperforms and lacks tuned SM10x kernel path
- Issue comment#1129636580292026-05-15 05:05How long does the memory-mechanism need?
- Issue#109qqingzheng2026-05-08 07:05严正声明：网络不是法外之地！立即停止针对 Helios 仓库的一切恶意攻击
- Issue#105bladpbuke2026-05-08 04:50[反馈] 心疼那几个真实用户
- Issue comment#55rampanther7032026-05-06 06:28Is it possible to make the training data available?
- Issue comment#56stewart-jackson8132026-05-06 06:28Question about Helios-Base speed in Table 3
- Issue comment#60charlotte1988q502026-05-06 05:27About Self-Forcing / Deouple-DMD / SVI / Reward-Forcing
- Issue comment#66charlotte1988q502026-05-06 04:03meaning of length_bucket_options
- Issue comment#64charlotte1988q502026-05-06 04:03推理报错
- Issue#81realhouse2262026-05-05 20:42對該倉庫近期星標異常增長表示關切
- Issue comment#73stewart-jackson8132026-05-05 20:37Question about inference speed and VRAM used 推理速度及显存打印与实际不符问题
- Issue comment#42splinter212026-04-17 10:25stage1init的效果问题
- Issue comment#68SHYuanBest2026-04-17 04:19推理效果疑问
- Issue#68splinter212026-04-17 04:17推理效果疑问
- Issue comment#60skywalker000012026-04-13 08:04About Self-Forcing / Deouple-DMD / SVI / Reward-Forcing
- Issue comment#64SHYuanBest2026-04-13 02:36推理报错
- Issue comment#60skywalker000012026-04-10 09:07About Self-Forcing / Deouple-DMD / SVI
- Issue comment#62SHYuanBest2026-04-10 03:55请问有Ascend npu的推理文档吗
- Issue comment#62OrangePure2026-04-10 03:48请问有Ascend npu的推理文档吗
- Issue comment#62SHYuanBest2026-04-10 02:47请问有Ascend npu的推理文档吗
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 388 stars here means stars gained during the window, not the repo's star count.