# Visual Question Answering on SlakeVQA

Source: [Visual Question Answering on SlakeVQA](<https://www.sota2.com/research/sota/visual-question-answering-on-slakevqa>)

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

Data updated: 2026-04-08T07:48:46.909Z

Evaluation of Visual Question Answering performance on the SlakeVQA dataset, reporting Tokenized F1 scores.

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 | Visual question answering |
| Dataset (including any stated split) | SlakeVQA |
| Metric | Tokenized F1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [Tokenized F1](<https://www.sota2.com/research/sota/visual-question-answering-on-slakevqa.md?metric=Tokenized%20F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Tokenized F1 | Result date | Report paper |
| --- | --- | --- | --- |
| MedGemma 1 | 72.3 | Not recorded | 69d5e1237d0172140ec75b36 |
| MedGemma 1 | 70.3 | Not recorded | 69d5e1237d0172140ec75b36 |
| Gemini 3 Pro | 62.8 | Not recorded | 69d5e1237d0172140ec75b36 |
| Gemini 3 Flash | 62.5 | Not recorded | 69d5e1237d0172140ec75b36 |
| MedGemma 1.5 | 59.8 | Not recorded | 69d5e1237d0172140ec75b36 |
| Qwen3 VL | 53.5 | Not recorded | 69d5e1237d0172140ec75b36 |

### MedGemma 1

Tokenized F1: 72.3. Result ID: 69d6085547f34ce924be5487.

[Reporting paper](<https://www.sota2.com/research/paper/69d5e1237d0172140ec75b36.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ff1e209131cea808a4.md>)

Recorded evaluation configuration: {"Model Scale":"4B","Temperature":"0.0"}. This record does not assert independent reproduction.

### MedGemma 1

Tokenized F1: 70.3. Result ID: 69d6085547f34ce924be5489.

[Reporting paper](<https://www.sota2.com/research/paper/69d5e1237d0172140ec75b36.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ff1e209131cea808a4.md>)

Recorded evaluation configuration: {"Model Scale":"27B","Temperature":"0.0"}. This record does not assert independent reproduction.

### Gemini 3 Pro

Tokenized F1: 62.8. Result ID: 69d6085547f34ce924be548c.

[Reporting paper](<https://www.sota2.com/research/paper/69d5e1237d0172140ec75b36.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73141e209131cea5a5c0.md>)

Recorded evaluation configuration: {"Temperature":"default"}. This record does not assert independent reproduction.

### Gemini 3 Flash

Tokenized F1: 62.5. Result ID: 69d6085547f34ce924be548b.

[Reporting paper](<https://www.sota2.com/research/paper/69d5e1237d0172140ec75b36.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73141e209131cea5a5c0.md>)

Recorded evaluation configuration: {"Temperature":"default"}. This record does not assert independent reproduction.

### MedGemma 1.5

Tokenized F1: 59.8. Result ID: 69d6085547f34ce924be5488.

[Reporting paper](<https://www.sota2.com/research/paper/69d5e1237d0172140ec75b36.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d5e1237d0172140ec75b36.md>)

Recorded evaluation configuration: {"Model Scale":"4B","Temperature":"0.0"}. This record does not assert independent reproduction.

### Qwen3 VL

Tokenized F1: 53.5. Result ID: 69d6085547f34ce924be548a.

[Reporting paper](<https://www.sota2.com/research/paper/69d5e1237d0172140ec75b36.md>)

Recorded evaluation configuration: {"Model Scale":"4B","Temperature":"default"}. This record does not assert independent reproduction.

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