# Real-World Understanding on RealWorldQA, MME-RW, and R-Bench

Source: [Real-World Understanding on RealWorldQA, MME-RW, and R-Bench](<https://www.sota2.com/research/sota/real-world-understanding-on-realworldqa-mme-rw-and-r-bench>)

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

Data updated: 2026-02-26T06:01:09.419Z

Evaluation of model performance across the RealWorldQA, MME-RW, and R-Bench components for the Real-World Understanding task.

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 | Real-World Understanding |
| Dataset (including any stated split) | Real-world Understanding (RealWorldQA, MME-RW, R-Bench) |
| Metric | RealWorld QA Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [RealWorld QA Score](<https://www.sota2.com/research/sota/real-world-understanding-on-realworldqa-mme-rw-and-r-bench.md?metric=RealWorld%20QA%20Score>): Higher is better; unit: PERCENT.

- [MME-RW (EN) Score](<https://www.sota2.com/research/sota/real-world-understanding-on-realworldqa-mme-rw-and-r-bench.md?metric=MME-RW%20(EN)%20Score>): Higher is better; unit: PERCENT.

- [R-Bench (Discrete) Score](<https://www.sota2.com/research/sota/real-world-understanding-on-realworldqa-mme-rw-and-r-bench.md?metric=R-Bench%20(Discrete)%20Score>): Higher is better; unit: PERCENT.

- [Overall RealWorld Understanding Score](<https://www.sota2.com/research/sota/real-world-understanding-on-realworldqa-mme-rw-and-r-bench.md?metric=Overall%20RealWorld%20Understanding%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | RealWorld QA Score | Result date | Report paper |
| --- | --- | --- | --- |
| jina-vlm | 68.2 | Not recorded | 693115d7d6b0dd3342933d9a |
| InternVL3-2B | 64.3 | Not recorded | 693115d7d6b0dd3342933d9a |
| Qwen3-VL-2B | 63.9 | Not recorded | 693115d7d6b0dd3342933d9a |
| Qwen2-VL-2B | 62.9 | Not recorded | 693115d7d6b0dd3342933d9a |
| InternVL3.5-2B | 62 | Not recorded | 693115d7d6b0dd3342933d9a |

### jina-vlm

RealWorld QA Score: 68.2. Result ID: 6996ce165698665ec41a0f7a.

[Reporting paper](<https://www.sota2.com/research/paper/693115d7d6b0dd3342933d9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/693115d7d6b0dd3342933d9a.md>)

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

### InternVL3-2B

RealWorld QA Score: 64.3. Result ID: 6996ce165698665ec41a0f7d.

[Reporting paper](<https://www.sota2.com/research/paper/693115d7d6b0dd3342933d9a.md>)

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

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

### Qwen3-VL-2B

RealWorld QA Score: 63.9. Result ID: 6996ce165698665ec41a0f7c.

[Reporting paper](<https://www.sota2.com/research/paper/693115d7d6b0dd3342933d9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75151e209131cea65976.md>)

[Implementation](<https://github.com/QwenLM/Qwen2-VL>)

Recorded evaluation configuration: {"evaluation\_toolkit":"VLMEvalKit"}. This record does not assert independent reproduction.

### Qwen2-VL-2B

RealWorld QA Score: 62.9. Result ID: 6996ce165698665ec41a0f7b.

[Reporting paper](<https://www.sota2.com/research/paper/693115d7d6b0dd3342933d9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7af01e209131cea85dc9.md>)

Recorded evaluation configuration: {"evaluation\_toolkit":"VLMEvalKit"}. This record does not assert independent reproduction.

### InternVL3.5-2B

RealWorld QA Score: 62. Result ID: 6996ce165698665ec41a0f7e.

[Reporting paper](<https://www.sota2.com/research/paper/693115d7d6b0dd3342933d9a.md>)

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