# Video Understanding on VideoBench

Source: [Video Understanding on VideoBench](<https://www.sota2.com/research/sota/video-understanding-on-videobench>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:34:40.487Z

Evaluation of general video understanding capabilities using the VideoBench suite, aggregating performance across multiple standard video tasks.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Video Understanding |
| Dataset (including any stated split) | VideoBench |
| Metric | ActivityNet Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 11 of 11 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [ActivityNet Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=ActivityNet%20Score>): Higher is better; unit: PERCENT.

- [MSVD Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=MSVD%20Score>): Higher is better; unit: PERCENT.

- [MSRVTT Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=MSRVTT%20Score>): Higher is better; unit: PERCENT.

- [TGIF Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=TGIF%20Score>): Higher is better; unit: PERCENT.

- [YC2 Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=YC2%20Score>): Higher is better; unit: PERCENT.

- [UCF Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=UCF%20Score>): Higher is better; unit: PERCENT.

- [MOT Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=MOT%20Score>): Higher is better; unit: PERCENT.

- [TV Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=TV%20Score>): Higher is better; unit: PERCENT.

- [MV Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=MV%20Score>): Higher is better; unit: PERCENT.

- [NBA Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=NBA%20Score>): Higher is better; unit: PERCENT.

- [LE Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=LE%20Score>): Higher is better; unit: PERCENT.

- [DM Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=DM%20Score>): Higher is better; unit: PERCENT.

- [SQA3D Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=SQA3D%20Score>): Higher is better; unit: PERCENT.

- [Average Score](<https://www.sota2.com/research/sota/video-understanding-on-videobench.md?metric=Average%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | ActivityNet Score | Result date | Report paper |
| --- | --- | --- | --- |
| ShareGPT4Video-8B | 50.8 | Not recorded | 68fa738e1e209131cea5d310 |
| ChatUniVi-7B | 49 | Not recorded | 68fa738e1e209131cea5d310 |
| VideoChatGPT-7B | 46.6 | Not recorded | 68fa738e1e209131cea5d310 |
| LLaMA-VID-7B | 45.2 | Not recorded | 68fa738e1e209131cea5d310 |
| PandaGPT-7B | 45 | Not recorded | 68fa738e1e209131cea5d310 |
| VideoChat-7B | 44.6 | Not recorded | 68fa738e1e209131cea5d310 |
| Otter-7B | 44.3 | Not recorded | 68fa738e1e209131cea5d310 |
| VideoLLaVA-7B | 44.1 | Not recorded | 68fa738e1e209131cea5d310 |
| mPLUG-Owl-7B | 41.5 | Not recorded | 68fa738e1e209131cea5d310 |
| Video-LLaMA-7B | 39.9 | Not recorded | 68fa738e1e209131cea5d310 |
| Valley-7B | 38.1 | Not recorded | 68fa738e1e209131cea5d310 |

### ShareGPT4Video-8B

ActivityNet Score: 50.8. Result ID: 6992ffcc3a6c31484912f338.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Implementation](<https://sharegpt4video.github.io/>)

Recorded evaluation configuration: {"number of parameters":"8B"}. This record does not assert independent reproduction.

### ChatUniVi-7B

ActivityNet Score: 49. Result ID: 6992ffcc3a6c31484912f335.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d31e209131cea58bff.md>)

[Implementation](<https://github.com/PKU-YuanGroup/Chat-UniVi>)

Recorded evaluation configuration: {"number of parameters":"7B"}. This record does not assert independent reproduction.

### VideoChatGPT-7B

ActivityNet Score: 46.6. Result ID: 6992ffcc3a6c31484912f334.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73841e209131cea5ce32.md>)

[Implementation](<https://github.com/mbzuai-oryx/Video-ChatGPT>)

Recorded evaluation configuration: {"number of parameters":"7B"}. This record does not assert independent reproduction.

### LLaMA-VID-7B

ActivityNet Score: 45.2. Result ID: 6992ffcc3a6c31484912f337.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71461e209131cea50559.md>)

[Implementation](<https://github.com/dvlab-research/LLaMA-VID>)

Recorded evaluation configuration: {"number of parameters":"7B","evaluation\_source":"public checkpoints"}. This record does not assert independent reproduction.

### PandaGPT-7B

ActivityNet Score: 45. Result ID: 6992ffcc3a6c31484912f333.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72581e209131cea55cc0.md>)

Recorded evaluation configuration: {"number of parameters":"7B"}. This record does not assert independent reproduction.

### VideoChat-7B

ActivityNet Score: 44.6. Result ID: 6992ffcc3a6c31484912f332.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71ba1e209131cea52cc7.md>)

Recorded evaluation configuration: {"number of parameters":"7B"}. This record does not assert independent reproduction.

### Otter-7B

ActivityNet Score: 44.3. Result ID: 6992ffcc3a6c31484912f32f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a411e209131cea81f51.md>)

Recorded evaluation configuration: {"number of parameters":"7B"}. This record does not assert independent reproduction.

### VideoLLaVA-7B

ActivityNet Score: 44.1. Result ID: 6992ffcc3a6c31484912f336.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75051e209131cea65534.md>)

[Implementation](<https://github.com/PKU-YuanGroup/Video-LLaVA>)

Recorded evaluation configuration: {"number of parameters":"7B","evaluation\_source":"public checkpoints"}. This record does not assert independent reproduction.

### mPLUG-Owl-7B

ActivityNet Score: 41.5. Result ID: 6992ffcc3a6c31484912f32e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ae1e209131cea57d83.md>)

[Implementation](<https://github.com/X-PLUG/mPLUG-Owl>)

Recorded evaluation configuration: {"number of parameters":"7B"}. This record does not assert independent reproduction.

### Video-LLaMA-7B

ActivityNet Score: 39.9. Result ID: 6992ffcc3a6c31484912f330.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f11e209131cea4e393.md>)

[Implementation](<https://github.com/DAMO-NLP-SG/Video-LLaMA>)

Recorded evaluation configuration: {"number of parameters":"7B"}. This record does not assert independent reproduction.

### Valley-7B

ActivityNet Score: 38.1. Result ID: 6992ffcc3a6c31484912f331.

[Reporting paper](<https://www.sota2.com/research/paper/68fa738e1e209131cea5d310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa777a1e209131cea73712.md>)

Recorded evaluation configuration: {"number of parameters":"7B"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
