# Long-video sustained reasoning on LVOmniBench

Source: [Long-video sustained reasoning on LVOmniBench](<https://www.sota2.com/research/sota/long-video-sustained-reasoning-on-lvomnibench>)

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

Data updated: 2026-05-22T08:06:13.414Z

Evaluation of long-video sustained reasoning capabilities using the LVOmniBench dataset, measured by accuracy.

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 | Long-video sustained reasoning |
| Dataset (including any stated split) | LVOmniBench |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/long-video-sustained-reasoning-on-lvomnibench.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Gemini-2.0-Flash | 42.9 | Not recorded | 6a0fe328f984739ed6739ea8 |
| LatentOmni | 35.1 | Not recorded | 6a0fe328f984739ed6739ea8 |
| MiniCPM-o-7B | 34.8 | Not recorded | 6a0fe328f984739ed6739ea8 |
| Qwen2.5-Omni-7B + Vanilla SFT | 33.2 | Not recorded | 6a0fe328f984739ed6739ea8 |
| Baichuan-Omni-1.5 | 32.8 | Not recorded | 6a0fe328f984739ed6739ea8 |
| HumanOmniV2-7B | 32.3 | Not recorded | 6a0fe328f984739ed6739ea8 |
| Qwen2.5-Omni-7B + Explicit Text CoT | 32.1 | Not recorded | 6a0fe328f984739ed6739ea8 |
| Qwen2.5-Omni-7B | 32 | Not recorded | 6a0fe328f984739ed6739ea8 |
| VideoLLaMA2-7B | 27 | Not recorded | 6a0fe328f984739ed6739ea8 |

### Gemini-2.0-Flash

Accuracy: 42.9. Result ID: 6a100e71fbc373592c08daf4.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

Recorded evaluation configuration: {"Model Accessibility":"Proprietary"}. This record does not assert independent reproduction.

### LatentOmni

Accuracy: 35.1. Result ID: 6a100e71fbc373592c08daf3.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

[Implementation](<https://github.com/yfanDai/LatentOmni>)

Recorded evaluation configuration: {"Model Accessibility":"Open-source","Training/Reasoning Strategy":"Latent reasoning"}. This record does not assert independent reproduction.

### MiniCPM-o-7B

Accuracy: 34.8. Result ID: 6a100e71fbc373592c08daed.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74681e209131cea6183d.md>)

[Implementation](<https://github.com/OpenBMB/MiniCPM-V>)

Recorded evaluation configuration: {"Model Accessibility":"Open-source"}. This record does not assert independent reproduction.

### Qwen2.5-Omni-7B + Vanilla SFT

Accuracy: 33.2. Result ID: 6a100e71fbc373592c08daf2.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

Recorded evaluation configuration: {"Model Accessibility":"Open-source","Training/Reasoning Strategy":"Vanilla SFT"}. This record does not assert independent reproduction.

### Baichuan-Omni-1.5

Accuracy: 32.8. Result ID: 6a100e71fbc373592c08daef.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

Recorded evaluation configuration: {"Model Accessibility":"Open-source"}. This record does not assert independent reproduction.

### HumanOmniV2-7B

Accuracy: 32.3. Result ID: 6a100e71fbc373592c08daee.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79a41e209131cea7ea02.md>)

Recorded evaluation configuration: {"Model Accessibility":"Open-source"}. This record does not assert independent reproduction.

### Qwen2.5-Omni-7B + Explicit Text CoT

Accuracy: 32.1. Result ID: 6a100e71fbc373592c08daf1.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

Recorded evaluation configuration: {"Model Accessibility":"Open-source","Training/Reasoning Strategy":"Explicit Text CoT"}. This record does not assert independent reproduction.

### Qwen2.5-Omni-7B

Accuracy: 32. Result ID: 6a100e71fbc373592c08daf0.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a21e209131cea7482c.md>)

Recorded evaluation configuration: {"Model Accessibility":"Open-source","Training/Reasoning Strategy":"Base"}. This record does not assert independent reproduction.

### VideoLLaMA2-7B

Accuracy: 27. Result ID: 6a100e71fbc373592c08daec.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe328f984739ed6739ea8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75761e209131cea681dd.md>)

[Implementation](<https://github.com/DAMO-NLP-SG/VideoLLaMA2>)

Recorded evaluation configuration: {"Model Accessibility":"Open-source"}. This record does not assert independent reproduction.

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