# Video Question Answering on Video-MME no subtitles v2 (test)

Source: [Video Question Answering on Video-MME no subtitles v2 (test)](<https://www.sota2.com/research/sota/video-question-answering-on-video-mme-no-subtitles-v2-test>)

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

Data updated: 2026-05-12T16:44:24.795Z

Evaluation of Video Question Answering performance on the test split of the Video-MME no subtitles v2 dataset, reporting accuracy metrics (Non-Linear, Average) and computational cost (TFLOPs).

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 Question Answering |
| Dataset (including any stated split) | Video-MME no subtitles v2 (test) |
| Metric | Non-Linear 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

- [Non-Linear Accuracy](<https://www.sota2.com/research/sota/video-question-answering-on-video-mme-no-subtitles-v2-test.md?metric=Non-Linear%20Accuracy>): Higher is better; unit: PERCENT.

- [Average Accuracy](<https://www.sota2.com/research/sota/video-question-answering-on-video-mme-no-subtitles-v2-test.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

- [TFLOPs](<https://www.sota2.com/research/sota/video-question-answering-on-video-mme-no-subtitles-v2-test.md?metric=TFLOPs>): Lower is better; unit: OTHER.

## Recorded results

| Method | Non-Linear Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen3-VL-8B | 14.91 | Not recorded | 6a02b4addbc836b87d824f22 |
| Qwen3-VL-4B | 14.11 | Not recorded | 6a02b4addbc836b87d824f22 |
| GP-4B | 12.86 | Not recorded | 6a02b4addbc836b87d824f22 |
| GP-8B | 12.6 | Not recorded | 6a02b4addbc836b87d824f22 |
| GP-2B-\>8B | 11.7 | Not recorded | 6a02b4addbc836b87d824f22 |
| Uniform-Meff -\> 8B | 10.83 | Not recorded | 6a02b4addbc836b87d824f22 |
| GP-2B-\>4B | 10.76 | Not recorded | 6a02b4addbc836b87d824f22 |
| Qwen3-VL-2B | 9.45 | Not recorded | 6a02b4addbc836b87d824f22 |
| GP-2B | 8.39 | Not recorded | 6a02b4addbc836b87d824f22 |

### Qwen3-VL-8B

Non-Linear Accuracy: 14.91. Result ID: 6a0358cf47f34ce924e8e6ed.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Pipeline":"Monolithic Baselines","K (Frame Pool Size)":"144","Model Size":"8B"}. This record does not assert independent reproduction.

### Qwen3-VL-4B

Non-Linear Accuracy: 14.11. Result ID: 6a0358cf47f34ce924e8e6ec.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Pipeline":"Monolithic Baselines","K (Frame Pool Size)":"144","Model Size":"4B"}. This record does not assert independent reproduction.

### GP-4B

Non-Linear Accuracy: 12.86. Result ID: 6a0358cf47f34ce924e8e6ef.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Implementation](<https://github.com/mohammad2012191/GridProbe>)

Recorded evaluation configuration: {"Pipeline":"Single-Model GridProbe","Selector Model":"Qwen3-VL-4B","QA Model":"Qwen3-VL-4B","K (Frame Pool Size)":"144"}. This record does not assert independent reproduction.

### GP-8B

Non-Linear Accuracy: 12.6. Result ID: 6a0358cf47f34ce924e8e6f0.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Implementation](<https://github.com/mohammad2012191/GridProbe>)

Recorded evaluation configuration: {"Pipeline":"Single-Model GridProbe","Selector Model":"Qwen3-VL-8B","QA Model":"Qwen3-VL-8B","K (Frame Pool Size)":"144"}. This record does not assert independent reproduction.

### GP-2B-\>8B

Non-Linear Accuracy: 11.7. Result ID: 6a0358cf47f34ce924e8e6f3.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Implementation](<https://github.com/mohammad2012191/GridProbe>)

Recorded evaluation configuration: {"Pipeline":"Cross-Model GridProbe","Selector Model":"Qwen3-VL-2B","QA Model":"Qwen3-VL-8B","K (Frame Pool Size)":"144"}. This record does not assert independent reproduction.

### Uniform-Meff -\> 8B

Non-Linear Accuracy: 10.83. Result ID: 6a0358cf47f34ce924e8e6f1.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Implementation](<https://github.com/mohammad2012191/GridProbe>)

Recorded evaluation configuration: {"Pipeline":"Cross-Model GridProbe (Control)","Selection":"Uniform","QA Model":"Qwen3-VL-8B","K (Frame Pool Size)":"144"}. This record does not assert independent reproduction.

### GP-2B-\>4B

Non-Linear Accuracy: 10.76. Result ID: 6a0358cf47f34ce924e8e6f2.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Implementation](<https://github.com/mohammad2012191/GridProbe>)

Recorded evaluation configuration: {"Pipeline":"Cross-Model GridProbe","Selector Model":"Qwen3-VL-2B","QA Model":"Qwen3-VL-4B","K (Frame Pool Size)":"144"}. This record does not assert independent reproduction.

### Qwen3-VL-2B

Non-Linear Accuracy: 9.45. Result ID: 6a0358cf47f34ce924e8e6eb.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Pipeline":"Monolithic Baselines","K (Frame Pool Size)":"144","Model Size":"2B"}. This record does not assert independent reproduction.

### GP-2B

Non-Linear Accuracy: 8.39. Result ID: 6a0358cf47f34ce924e8e6ee.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b4addbc836b87d824f22.md>)

[Implementation](<https://github.com/mohammad2012191/GridProbe>)

Recorded evaluation configuration: {"Pipeline":"Single-Model GridProbe","Selector Model":"Qwen3-VL-2B","QA Model":"Qwen3-VL-2B","K (Frame Pool Size)":"144"}. This record does not assert independent reproduction.

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