# Long-Video Understanding on VideoMME (1-60m, w/o subtitles)

Source: [Long-Video Understanding on VideoMME (1-60m, w/o subtitles)](<https://www.sota2.com/research/sota/long-video-understanding-on-videomme-1-60m-w-o-subtitles>)

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

Data updated: 2026-03-04T09:24:21.867Z

Evaluation of Long-Video Understanding capabilities on the VideoMME dataset subset covering video lengths from 1 to 60 minutes, specifically excluding subtitles, measured by the generic Score metric.

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 Understanding |
| Dataset (including any stated split) | VideoMME 1~60m, w/o subtitles |
| Metric | Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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

- [Score](<https://www.sota2.com/research/sota/long-video-understanding-on-videomme-1-60m-w-o-subtitles.md?metric=Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Score | Result date | Report paper |
| --- | --- | --- | --- |
| Gemini-1.5-Pro | 75 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| Qwen2.5+FlexSelect | 74.4 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| GPT-4o | 71.9 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| Nemotron-Nano-V2 VL + All layer reduction | 69.7 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| LLaVA-Video+FlexSelect | 68.9 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| Qwen3 VL + All layer reduction | 68.52 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| Qwen2.5 VL+FlexSelect | 68.2 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| ViLAMP | 67.5 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| Oryx-1.5 | 67.3 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| Qwen2.5 VL+SparseVILA | 66.3 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| VideoLLaMA3 | 66.2 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| NVILA | 64.2 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| SF-LLaVA-1.5 | 63.9 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| Qwen2-VL | 63.3 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| Apollo | 61.3 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| LongVU | 60.6 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| MLLM-FS | 58.7 | Not recorded | 69a66bdf2d603e6de3e3ccfd |
| LLaVA-Onevision | 58.2 | Not recorded | 69a66bdf2d603e6de3e3ccfd |

### Gemini-1.5-Pro

Score: 75. Result ID: 69a7f8e86dbb7075b94d1104.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78731e209131cea78622.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"–"}. This record does not assert independent reproduction.

### Qwen2.5+FlexSelect

Score: 74.4. Result ID: 69a7f8e86dbb7075b94d1112.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78ff1e209131cea7b8e1.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"72B"}. This record does not assert independent reproduction.

### GPT-4o

Score: 71.9. Result ID: 69a7f8e86dbb7075b94d1103.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75681e209131cea67a98.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"–"}. This record does not assert independent reproduction.

### Nemotron-Nano-V2 VL + All layer reduction

Score: 69.7. Result ID: 69a7f8e86dbb7075b94d1114.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

Recorded evaluation configuration: {"Model Type":"Hybrid","Size":"12B"}. This record does not assert independent reproduction.

### LLaVA-Video+FlexSelect

Score: 68.9. Result ID: 69a7f8e86dbb7075b94d1111.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78ff1e209131cea7b8e1.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### Qwen3 VL + All layer reduction

Score: 68.52. Result ID: 69a7f8e86dbb7075b94d1113.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"8B"}. This record does not assert independent reproduction.

### Qwen2.5 VL+FlexSelect

Score: 68.2. Result ID: 69a7f8e86dbb7075b94d1110.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78ff1e209131cea7b8e1.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### ViLAMP

Score: 67.5. Result ID: 69a7f8e86dbb7075b94d110e.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b3e1e209131cea87432.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### Oryx-1.5

Score: 67.3. Result ID: 69a7f8e86dbb7075b94d110a.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa772c1e209131cea71918.md>)

[Implementation](<https://github.com/Oryx-mllm/Oryx>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"34B"}. This record does not assert independent reproduction.

### Qwen2.5 VL+SparseVILA

Score: 66.3. Result ID: 69a7f8e86dbb7075b94d110f.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c5e1e209131cea8dc56.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### VideoLLaMA3

Score: 66.2. Result ID: 69a7f8e86dbb7075b94d1109.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79111e209131cea7bf26.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### NVILA

Score: 64.2. Result ID: 69a7f8e86dbb7075b94d1107.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77481e209131cea723b2.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"8B"}. This record does not assert independent reproduction.

### SF-LLaVA-1.5

Score: 63.9. Result ID: 69a7f8e86dbb7075b94d110d.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779a1e209131cea74494.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### Qwen2-VL

Score: 63.3. Result ID: 69a7f8e86dbb7075b94d1106.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75151e209131cea65976.md>)

[Implementation](<https://github.com/QwenLM/Qwen2-VL>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### Apollo

Score: 61.3. Result ID: 69a7f8e86dbb7075b94d1108.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa763b1e209131cea6c018.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### LongVU

Score: 60.6. Result ID: 69a7f8e86dbb7075b94d110c.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75511e209131cea671f3.md>)

[Implementation](<https://github.com/Vision-CAIR/LongVU>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### MLLM-FS

Score: 58.7. Result ID: 69a7f8e86dbb7075b94d110b.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a51e209131cea74983.md>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

### LLaVA-Onevision

Score: 58.2. Result ID: 69a7f8e86dbb7075b94d1105.

[Reporting paper](<https://www.sota2.com/research/paper/69a66bdf2d603e6de3e3ccfd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756f1e209131cea67e58.md>)

[Implementation](<https://llava-vl.github.io/blog/llava-onevision>)

Recorded evaluation configuration: {"Model Type":"Transformer","Size":"7B"}. This record does not assert independent reproduction.

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