# Open-ended Question Answering on MMAU-Pro Open-ended

Source: [Open-ended Question Answering on MMAU-Pro Open-ended](<https://www.sota2.com/research/sota/open-ended-question-answering-on-mmau-pro-open-ended>)

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

Data updated: 2026-05-20T09:04:10.918Z

Evaluation of model performance on the MMAU-Pro Open-ended dataset for Open-ended Question Answering, reporting Open-ended success rate and Good Rate.

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 | Open-ended Question Answering |
| Dataset (including any stated split) | MMAU-Pro Open-ended |
| Metric | Open-ended Success Rate |
| 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

- [Open-ended Success Rate](<https://www.sota2.com/research/sota/open-ended-question-answering-on-mmau-pro-open-ended.md?metric=Open-ended%20Success%20Rate>): Higher is better; unit: PERCENT.

- [Good Rate](<https://www.sota2.com/research/sota/open-ended-question-answering-on-mmau-pro-open-ended.md?metric=Good%20Rate>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Open-ended Success Rate | Result date | Report paper |
| --- | --- | --- | --- |
| OScaR | 67.4 | Not recorded | 6a0d4051f984739ed6737b03 |
| TurboQuant+ | 66.6 | Not recorded | 6a0d4051f984739ed6737b03 |
| 16bit baseline | 66.2 | Not recorded | 6a0d4051f984739ed6737b03 |
| KIVI | 65.8 | Not recorded | 6a0d4051f984739ed6737b03 |
| OTT | 65.8 | Not recorded | 6a0d4051f984739ed6737b03 |

### OScaR

Open-ended Success Rate: 67.4. Result ID: 6a0d78d4fbc373592c05afcd.

[Reporting paper](<https://www.sota2.com/research/paper/6a0d4051f984739ed6737b03.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0d4051f984739ed6737b03.md>)

[Implementation](<https://github.com/ZunhaiSu/OScaR-KV-Quant>)

Recorded evaluation configuration: {"Backbone":"Qwen3-Omni-30B-A3B","Quantization Method":"INT2","Group Size":"128","Bit-width":"INT2"}. This record does not assert independent reproduction.

### TurboQuant+

Open-ended Success Rate: 66.6. Result ID: 6a0d78d4fbc373592c05afcc.

[Reporting paper](<https://www.sota2.com/research/paper/6a0d4051f984739ed6737b03.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-Omni-30B-A3B","Bit-width":"2.5-bit"}. This record does not assert independent reproduction.

### 16bit baseline

Open-ended Success Rate: 66.2. Result ID: 6a0d78d4fbc373592c05afc9.

[Reporting paper](<https://www.sota2.com/research/paper/6a0d4051f984739ed6737b03.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-Omni-30B-A3B","Bit-width":"16-bit"}. This record does not assert independent reproduction.

### KIVI

Open-ended Success Rate: 65.8. Result ID: 6a0d78d4fbc373592c05afca.

[Reporting paper](<https://www.sota2.com/research/paper/6a0d4051f984739ed6737b03.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa744b1e209131cea60dab.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-Omni-30B-A3B","Quantization Method":"INT2","Group Size":"128","Bit-width":"INT2"}. This record does not assert independent reproduction.

### OTT

Open-ended Success Rate: 65.8. Result ID: 6a0d78d4fbc373592c05afcb.

[Reporting paper](<https://www.sota2.com/research/paper/6a0d4051f984739ed6737b03.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa789e1e209131cea7961c.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-Omni-30B-A3B","Quantization Method":"INT2","Group Size":"128","Bit-width":"INT2"}. This record does not assert independent reproduction.

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