# High-resolution Visual Question Answering on HR-Bench 8K

Source: [High-resolution Visual Question Answering on HR-Bench 8K](<https://www.sota2.com/research/sota/high-resolution-visual-question-answering-on-hr-bench-8k>)

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

Data updated: 2026-05-05T06:54:48.179Z

Evaluation of model performance on the HR-Bench 8K dataset specifically for High-resolution Visual Question Answering tasks, reporting Overall, Single, and Cross category 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 | High-resolution Visual Question Answering |
| Dataset (including any stated split) | HR-Bench 8K |
| Metric | Overall Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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

- [Overall Accuracy](<https://www.sota2.com/research/sota/high-resolution-visual-question-answering-on-hr-bench-8k.md?metric=Overall%20Accuracy>): Higher is better; unit: PERCENT.

- [Single Category Accuracy](<https://www.sota2.com/research/sota/high-resolution-visual-question-answering-on-hr-bench-8k.md?metric=Single%20Category%20Accuracy>): Higher is better; unit: PERCENT.

- [Cross Category Accuracy](<https://www.sota2.com/research/sota/high-resolution-visual-question-answering-on-hr-bench-8k.md?metric=Cross%20Category%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Overall Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| PFlowNet | 76.9 | Not recorded | 69f979fbc0f56ed843089c95 |
| Qwen2.5-VL-72B | 76.3 | Not recorded | 69f979fbc0f56ed843089c95 |
| Qwen3-VL-32B | 74.8 | Not recorded | 69f979fbc0f56ed843089c95 |
| DeepEyesV2 | 73.8 | Not recorded | 69f979fbc0f56ed843089c95 |
| TreeVGR | 72.8 | Not recorded | 69f979fbc0f56ed843089c95 |
| DeepEyes | 72.6 | Not recorded | 69f979fbc0f56ed843089c95 |
| Thyme | 72 | Not recorded | 69f979fbc0f56ed843089c95 |
| Qwen2.5-VL-32B | 71.6 | Not recorded | 69f979fbc0f56ed843089c95 |
| Qwen2.5-VL-7B | 68.8 | Not recorded | 69f979fbc0f56ed843089c95 |
| Qwen3-VL-8B | 68.1 | Not recorded | 69f979fbc0f56ed843089c95 |
| Qwen3-VL-4B | 67.1 | Not recorded | 69f979fbc0f56ed843089c95 |
| InternVL3-38B | 67 | Not recorded | 69f979fbc0f56ed843089c95 |
| PixelReasoner | 66.9 | Not recorded | 69f979fbc0f56ed843089c95 |
| InternVL3-8B | 62 | Not recorded | 69f979fbc0f56ed843089c95 |
| LLaVA-OV-72B | 60.9 | Not recorded | 69f979fbc0f56ed843089c95 |
| GPT-4o-1120 | 55.5 | Not recorded | 69f979fbc0f56ed843089c95 |

### PFlowNet

Overall Accuracy: 76.9. Result ID: 69f993c347f34ce924dcc073.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen2.5-VL-72B

Overall Accuracy: 76.3. Result ID: 69f993c347f34ce924dcc06a.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen3-VL-32B

Overall Accuracy: 74.8. Result ID: 69f993c347f34ce924dcc06d.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DeepEyesV2

Overall Accuracy: 73.8. Result ID: 69f993c347f34ce924dcc070.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691171652828b7cc6937d492.md>)

[Implementation](<https://github.com/Visual-Agent/DeepEyes>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TreeVGR

Overall Accuracy: 72.8. Result ID: 69f993c347f34ce924dcc072.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79fd1e209131cea807a5.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DeepEyes

Overall Accuracy: 72.6. Result ID: 69f993c347f34ce924dcc06f.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78c71e209131cea7a75e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Thyme

Overall Accuracy: 72. Result ID: 69f993c347f34ce924dcc071.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7aa21e209131cea84434.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen2.5-VL-32B

Overall Accuracy: 71.6. Result ID: 69f993c347f34ce924dcc069.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen2.5-VL-7B

Overall Accuracy: 68.8. Result ID: 69f993c347f34ce924dcc068.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen3-VL-8B

Overall Accuracy: 68.1. Result ID: 69f993c347f34ce924dcc06c.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen3-VL-4B

Overall Accuracy: 67.1. Result ID: 69f993c347f34ce924dcc06b.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### InternVL3-38B

Overall Accuracy: 67. Result ID: 69f993c347f34ce924dcc067.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78241e209131cea76fa5.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### PixelReasoner

Overall Accuracy: 66.9. Result ID: 69f993c347f34ce924dcc06e.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78c71e209131cea7a774.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### InternVL3-8B

Overall Accuracy: 62. Result ID: 69f993c347f34ce924dcc066.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78241e209131cea76fa5.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### LLaVA-OV-72B

Overall Accuracy: 60.9. Result ID: 69f993c347f34ce924dcc065.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

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

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

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GPT-4o-1120

Overall Accuracy: 55.5. Result ID: 69f993c347f34ce924dcc064.

[Reporting paper](<https://www.sota2.com/research/paper/69f979fbc0f56ed843089c95.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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