# Mathematical Reasoning on MathVista (GEO, ARI, GPS, ALL)

Source: [Mathematical Reasoning on MathVista (GEO, ARI, GPS, ALL)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-mathvista-geo-ari-gps-all>)

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

Data updated: 2026-06-25T07:54:16.088Z

Evaluation of model performance on the MathVista dataset for Mathematical Reasoning, reporting specific metrics including GEO, ARI, GPS, and the overall ALL 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 | Mathematical Reasoning |
| Dataset (including any stated split) | MathVista |
| Metric | Overall Score (ALL) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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 Score (ALL)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-mathvista-geo-ari-gps-all.md?metric=Overall%20Score%20(ALL)>): Higher is better; unit: PERCENT.

- [GEO Score](<https://www.sota2.com/research/sota/mathematical-reasoning-on-mathvista-geo-ari-gps-all.md?metric=GEO%20Score>): Higher is better; unit: PERCENT.

- [ARI Score](<https://www.sota2.com/research/sota/mathematical-reasoning-on-mathvista-geo-ari-gps-all.md?metric=ARI%20Score>): Higher is better; unit: PERCENT.

- [GPS Score](<https://www.sota2.com/research/sota/mathematical-reasoning-on-mathvista-geo-ari-gps-all.md?metric=GPS%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Overall Score (ALL) | Result date | Report paper |
| --- | --- | --- | --- |
| R1-Onevision-7B | 64.1 | Not recorded | 68fa7a4d1e209131cea82516 |
| GPT-4o | 63.8 | Not recorded | 68fa7a4d1e209131cea82516 |
| R1-VL-7B | 63.6 | Not recorded | 68fa7a4d1e209131cea82516 |
| MGT-PerceReason | 63.2 | Not recorded | 68fa7a4d1e209131cea82516 |
| SyncLoop-3rd | 63.2 | Not recorded | 68fa7a4d1e209131cea82516 |
| SyncLoop-2nd | 62.4 | Not recorded | 68fa7a4d1e209131cea82516 |
| SyncLoop-1st | 62.1 | Not recorded | 68fa7a4d1e209131cea82516 |
| Qwen2-VL-7B | 60 | Not recorded | 68fa7a4d1e209131cea82516 |
| LLamaV-o1-11B | 54.4 | Not recorded | 68fa7a4d1e209131cea82516 |
| SyncLoop-3rd | 50.2 | Not recorded | 68fa7a4d1e209131cea82516 |
| Insight-V-8B | 49.8 | Not recorded | 68fa7a4d1e209131cea82516 |
| SyncLoop-2nd | 49.3 | Not recorded | 68fa7a4d1e209131cea82516 |
| SyncLoop-1st | 49.1 | Not recorded | 68fa7a4d1e209131cea82516 |
| Qwen2-VL-2B | 43 | Not recorded | 68fa7a4d1e209131cea82516 |

### R1-Onevision-7B

Overall Score (ALL): 64.1. Result ID: 6a3cde96fbc373592c51cd33.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77771e209131cea7357a.md>)

Recorded evaluation configuration: {"Data Amount":"10k"}. This record does not assert independent reproduction.

### GPT-4o

Overall Score (ALL): 63.8. Result ID: 6a3cde96fbc373592c51cd2f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75681e209131cea67a98.md>)

Recorded evaluation configuration: {"Data Amount":"-"}. This record does not assert independent reproduction.

### R1-VL-7B

Overall Score (ALL): 63.6. Result ID: 6a3cde96fbc373592c51cd34.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a651e209131cea82e9b.md>)

Recorded evaluation configuration: {"Data Amount":"10k (SFT 120k)"}. This record does not assert independent reproduction.

### MGT-PerceReason

Overall Score (ALL): 63.2. Result ID: 6a3cde96fbc373592c51cd32.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

Recorded evaluation configuration: {"Data Amount":"65k"}. This record does not assert independent reproduction.

### SyncLoop-3rd

Overall Score (ALL): 63.2. Result ID: 6a3cde96fbc373592c51cd3c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2-VL-7B","Data Amount":"0.4k"}. This record does not assert independent reproduction.

### SyncLoop-2nd

Overall Score (ALL): 62.4. Result ID: 6a3cde96fbc373592c51cd3b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2-VL-7B","Data Amount":"0.6k"}. This record does not assert independent reproduction.

### SyncLoop-1st

Overall Score (ALL): 62.1. Result ID: 6a3cde96fbc373592c51cd3a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2-VL-7B","Data Amount":"0.6k"}. This record does not assert independent reproduction.

### Qwen2-VL-7B

Overall Score (ALL): 60. Result ID: 6a3cde96fbc373592c51cd39.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

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

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

Recorded evaluation configuration: {"Data Amount":"-"}. This record does not assert independent reproduction.

### LLamaV-o1-11B

Overall Score (ALL): 54.4. Result ID: 6a3cde96fbc373592c51cd30.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76881e209131cea6d7bc.md>)

Recorded evaluation configuration: {"Data Amount":"\>100k"}. This record does not assert independent reproduction.

### SyncLoop-3rd

Overall Score (ALL): 50.2. Result ID: 6a3cde96fbc373592c51cd38.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2-VL-2B","Data Amount":"0.4k"}. This record does not assert independent reproduction.

### Insight-V-8B

Overall Score (ALL): 49.8. Result ID: 6a3cde96fbc373592c51cd31.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78541e209131cea77be2.md>)

Recorded evaluation configuration: {"Data Amount":"200K"}. This record does not assert independent reproduction.

### SyncLoop-2nd

Overall Score (ALL): 49.3. Result ID: 6a3cde96fbc373592c51cd37.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2-VL-2B","Data Amount":"0.6k"}. This record does not assert independent reproduction.

### SyncLoop-1st

Overall Score (ALL): 49.1. Result ID: 6a3cde96fbc373592c51cd36.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2-VL-2B","Data Amount":"0.6k"}. This record does not assert independent reproduction.

### Qwen2-VL-2B

Overall Score (ALL): 43. Result ID: 6a3cde96fbc373592c51cd35.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a4d1e209131cea82516.md>)

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

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

Recorded evaluation configuration: {"Data Amount":"-"}. This record does not assert independent reproduction.

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