VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling
active 2025-01-02 → 2026-05-15 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 499 days from 2025-01-02 to 2026-05-15. Pushes: 34 total, peak 6 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 69 total, peak 5 in a day. Comments: 99 total, peak 8 in a day. Stars: 145 total, peak 6 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 |
|---|---|---|---|---|
| leexinhao | 111 | 34 | 0 | 54 |
| jun0wanan | 6 | 0 | 0 | 4 |
| lixinghe1999 | 4 | 0 | 0 | 3 |
| junia3 | 4 | 0 | 0 | 3 |
| Sakura-not-sleep | 4 | 0 | 0 | 2 |
| hello-bluedog | 4 | 0 | 0 | 1 |
| BoweiPu | 3 | 0 | 0 | 1 |
| ForHumble | 3 | 0 | 0 | 1 |
| ilileun | 3 | 0 | 0 | 1 |
| JonnesLin | 3 | 0 | 0 | 0 |
| ZheyuAqaZhang | 3 | 0 | 0 | 0 |
| ziqipang | 3 | 0 | 0 | 2 |
| UnableToUseGit | 3 | 0 | 0 | 1 |
| rishabh-akridata | 2 | 0 | 0 | 0 |
| MengHao666 | 2 | 0 | 0 | 1 |
| Noctis-SC | 2 | 0 | 0 | 1 |
| Jays-1111 | 2 | 0 | 0 | 0 |
| Columbine21 | 2 | 0 | 0 | 0 |
| lpy523 | 2 | 0 | 0 | 1 |
| Dave-he | 2 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue#74Jays-11112025-12-05 09:58有提供AMD GPU版的pytorch 推理代码么，当前安装flash attention会报错，导致无法推理
- Issue#73Jays-11112025-12-05 07:15VideoChat-Flash-Qwen2_5-7B_InternVideo2-1B windows上可以在哪个推理框架推理呢
- Issue#72jun0yayay2025-11-02 10:20baseline is qwen2.5vl
- Issue comment#26Arsiuuu2025-10-17 04:33fintune code error
- Issue#53leexinhao2025-09-15 08:56Single-device single-card torchrun fine-tuning
- Issue#71leexinhao2025-09-15 08:54MLVU评测问题
- Issue#42leexinhao2025-08-07 14:43petrel_client error when running the example code
- Issue comment#66leexinhao2025-08-07 14:41support for vLLM deployment
- Issue#70ForHumble2025-08-03 08:41Issues related to cpu memory
- Issue comment#70ForHumble2025-08-03 08:22Issues related to cpu memory
- Issue#70ForHumble2025-08-01 14:29Issues related to cpu memory
- Issue#69ForHumble2025-07-25 05:05Some minor issues with visual encoder loading
- Issue comment#69ForHumble2025-07-25 05:05Some minor issues with visual encoder loading
- Issue comment#68leexinhao2025-07-25 03:32多卡训练tensor不匹配
- Issue comment#68william-ljz2025-07-25 02:04多卡训练tensor不匹配
- Issue comment#69leexinhao2025-07-24 12:45Some minor issues with visual encoder loading
- Issue comment#68leexinhao2025-07-24 12:38多卡训练tensor不匹配
- Issue comment#68leexinhao2025-07-24 12:37多卡训练tensor不匹配
- Issue#69ForHumble2025-07-24 10:55Some minor issues with visual encoder loading
- Issue comment#68william-ljz2025-07-23 06:09多卡训练tensor不匹配
- Issue comment#68william-ljz2025-07-23 06:05多卡训练tensor不匹配
- Issue comment#68leexinhao2025-07-23 06:05多卡训练tensor不匹配
- Issue comment#68leexinhao2025-07-23 06:03多卡训练tensor不匹配
- Issue comment#68william-ljz2025-07-21 17:49多卡训练tensor不匹配
- Issue comment#68william-ljz2025-07-21 07:06多卡训练tensor不匹配
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 145 stars here means stars gained during the window, not the repo's star count.