# Face Recognition on YouTube Faces (test)

Source: [Face Recognition on YouTube Faces (test)](<https://www.sota2.com/research/sota/face-recognition-on-youtube-faces-test>)

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

Data updated: 2026-04-06T07:15:01.872Z

Evaluation of Face Recognition performance on the test split of the YouTube Faces dataset.

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 | Face Recognition |
| Dataset (including any stated split) | YouTube Faces (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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/face-recognition-on-youtube-faces-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Accuracy (w/ FaVChat)](<https://www.sota2.com/research/sota/face-recognition-on-youtube-faces-test.md?metric=Accuracy%20(w%2F%20FaVChat)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FaVChat | 88.38 | Not recorded | 68fa77611e209131cea72d93 |
| Claude4-Sonnet | 73.41 | Not recorded | 68fa77611e209131cea72d93 |
| GPT-4o | 71.29 | Not recorded | 68fa77611e209131cea72d93 |
| Qwen2.5-VL-72B | 66.84 | Not recorded | 68fa77611e209131cea72d93 |
| Qwen2.5-VL-7B-Face | 64.28 | Not recorded | 68fa77611e209131cea72d93 |
| VideoChat2 | 57.3 | Not recorded | 68fa77611e209131cea72d93 |
| VideoChat | 54.74 | Not recorded | 68fa77611e209131cea72d93 |
| VideoLLaMa2 | 51.03 | Not recorded | 68fa77611e209131cea72d93 |

### FaVChat

Accuracy: 88.38. Result ID: 69d35d2b47f34ce924bab71d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77611e209131cea72d93.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77611e209131cea72d93.md>)

Recorded evaluation configuration: {"Size":"7B","Zero-shot":"true"}. This record does not assert independent reproduction.

### Claude4-Sonnet

Accuracy: 73.41. Result ID: 69d35d2b47f34ce924bab717.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77611e209131cea72d93.md>)

Recorded evaluation configuration: {"Size":"-","Zero-shot":"true"}. This record does not assert independent reproduction.

### GPT-4o

Accuracy: 71.29. Result ID: 69d35d2b47f34ce924bab716.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77611e209131cea72d93.md>)

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

Recorded evaluation configuration: {"Size":"-","Zero-shot":"true"}. This record does not assert independent reproduction.

### Qwen2.5-VL-72B

Accuracy: 66.84. Result ID: 69d35d2b47f34ce924bab71c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77611e209131cea72d93.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Size":"72B","Zero-shot":"true"}. This record does not assert independent reproduction.

### Qwen2.5-VL-7B-Face

Accuracy: 64.28. Result ID: 69d35d2b47f34ce924bab71b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77611e209131cea72d93.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa773a1e209131cea71d3c.md>)

Recorded evaluation configuration: {"Size":"7B","Zero-shot":"true"}. This record does not assert independent reproduction.

### VideoChat2

Accuracy: 57.3. Result ID: 69d35d2b47f34ce924bab719.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77611e209131cea72d93.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73411e209131cea5b4b7.md>)

Recorded evaluation configuration: {"Size":"7B","Zero-shot":"true"}. This record does not assert independent reproduction.

### VideoChat

Accuracy: 54.74. Result ID: 69d35d2b47f34ce924bab718.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77611e209131cea72d93.md>)

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

Recorded evaluation configuration: {"Size":"7B","Zero-shot":"true"}. This record does not assert independent reproduction.

### VideoLLaMa2

Accuracy: 51.03. Result ID: 69d35d2b47f34ce924bab71a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77611e209131cea72d93.md>)

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

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

Recorded evaluation configuration: {"Size":"7B","Zero-shot":"true"}. This record does not assert independent reproduction.

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