# Visual Question Answering on LLaVA-Bench LLaVAW

Source: [Visual Question Answering on LLaVA-Bench LLaVAW](<https://www.sota2.com/research/sota/visual-question-answering-on-llava-bench-llavaw>)

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

Data updated: 2026-02-26T06:03:46.049Z

Evaluation of model performance on the LLaVA-Bench LLaVAW dataset split for general Visual Question Answering capabilities, reporting a primary 'Score'.

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) | LLaVA-Bench LLaVAW |
| Metric | Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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/visual-question-answering-on-llava-bench-llavaw.md?metric=Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Score | Result date | Report paper |
| --- | --- | --- | --- |
| LLaVA-VT | 89.1 | Not recorded | 68fa73b21e209131cea5df74 |
| LLaVA-Cap | 88.9 | Not recorded | 68fa73b21e209131cea5df74 |
| LLaVA-SG | 86.8 | Not recorded | 68fa73b21e209131cea5df74 |
| LLaVA-1.5 (V-13B) | 86.7 | Not recorded | 68fa73b21e209131cea5df74 |
| LLaVA-DCap | 86.4 | Not recorded | 68fa73b21e209131cea5df74 |
| Vicuna-VT | 82.5 | Not recorded | 68fa73b21e209131cea5df74 |
| LLaVA-1.5 (V-7B) | 81.9 | Not recorded | 68fa73b21e209131cea5df74 |
| Vicuna-Cap | 79.2 | Not recorded | 68fa73b21e209131cea5df74 |
| Vicuna-DCap | 77.6 | Not recorded | 68fa73b21e209131cea5df74 |
| Vicuna-SG | 77 | Not recorded | 68fa73b21e209131cea5df74 |

### LLaVA-VT

Score: 89.1. Result ID: 699309c63a6c314849133f8d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Implementation](<https://github.com/LaVi-Lab/Visual-Table>)

Recorded evaluation configuration: {"LLM":"Vicuna-13B","#PT":"558K","#IT":"665K","Representation":"E + T"}. This record does not assert independent reproduction.

### LLaVA-Cap

Score: 88.9. Result ID: 699309c63a6c314849133f8a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea46fe6.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/blip2>)

Recorded evaluation configuration: {"LLM":"Vicuna-13B","#PT":"558K","#IT":"665K","Representation":"E + T"}. This record does not assert independent reproduction.

### LLaVA-SG

Score: 86.8. Result ID: 699309c63a6c314849133f8c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72be1e209131cea584e9.md>)

[Implementation](<https://github.com/chancharikmitra/CCoT>)

Recorded evaluation configuration: {"LLM":"Vicuna-13B","#PT":"558K","#IT":"665K","Representation":"E + T"}. This record does not assert independent reproduction.

### LLaVA-1.5 (V-13B)

Score: 86.7. Result ID: 699309c63a6c314849133f85.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af59.md>)

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

Recorded evaluation configuration: {"LLM":"Vicuna-13B","#PT":"558K","#IT":"665K","Representation":"Visual Embeddings (E)"}. This record does not assert independent reproduction.

### LLaVA-DCap

Score: 86.4. Result ID: 699309c63a6c314849133f8b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af59.md>)

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

Recorded evaluation configuration: {"LLM":"Vicuna-13B","#PT":"558K","#IT":"665K","Representation":"E + T"}. This record does not assert independent reproduction.

### Vicuna-VT

Score: 82.5. Result ID: 699309c63a6c314849133f89.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Implementation](<https://github.com/LaVi-Lab/Visual-Table>)

Recorded evaluation configuration: {"LLM":"Vicuna-13B","#IT":"665K","Representation":"Text (T)"}. This record does not assert independent reproduction.

### LLaVA-1.5 (V-7B)

Score: 81.9. Result ID: 699309c63a6c314849133f84.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af59.md>)

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

Recorded evaluation configuration: {"LLM":"Vicuna-7B","#PT":"558K","#IT":"665K","Representation":"Visual Embeddings (E)"}. This record does not assert independent reproduction.

### Vicuna-Cap

Score: 79.2. Result ID: 699309c63a6c314849133f86.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea46fe6.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/blip2>)

Recorded evaluation configuration: {"LLM":"Vicuna-13B","#IT":"665K","Representation":"Text (T)"}. This record does not assert independent reproduction.

### Vicuna-DCap

Score: 77.6. Result ID: 699309c63a6c314849133f87.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af59.md>)

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

Recorded evaluation configuration: {"LLM":"Vicuna-13B","#IT":"665K","Representation":"Text (T)"}. This record does not assert independent reproduction.

### Vicuna-SG

Score: 77. Result ID: 699309c63a6c314849133f88.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df74.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72be1e209131cea584e9.md>)

[Implementation](<https://github.com/chancharikmitra/CCoT>)

Recorded evaluation configuration: {"LLM":"Vicuna-13B","#IT":"665K","Representation":"Text (T)"}. This record does not assert independent reproduction.

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