# Mathematical Reasoning on FlowVerse Text Plus

Source: [Mathematical Reasoning on FlowVerse Text Plus](<https://www.sota2.com/research/sota/mathematical-reasoning-on-flowverse-text-plus>)

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

Data updated: 2026-04-21T08:41:57.008Z

Evaluation of model performance on the 'Text Plus' subset of the FlowVerse dataset for Mathematical Reasoning, measured using Accuracy (Acc) and Chain-of-Thought Error (CoT-E).

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) | FlowVerse Text Plus |
| Metric | CoT Error |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 15 of 15 |
| 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

- [CoT Error](<https://www.sota2.com/research/sota/mathematical-reasoning-on-flowverse-text-plus.md?metric=CoT%20Error>): Lower is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-flowverse-text-plus.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | CoT Error | Result date | Report paper |
| --- | --- | --- | --- |
| InfiMM-Math-7B | 46.1 | Not recorded | 68fa77951e209131cea741e7 |
| InternVL2.5-8B | 49.6 | Not recorded | 68fa77951e209131cea741e7 |
| Qwen-VL-Max | 53.9 | Not recorded | 68fa77951e209131cea741e7 |
| GPT-4o-mini | 59.6 | Not recorded | 68fa77951e209131cea741e7 |
| Qwen2.5-VL-7B | 62 | Not recorded | 68fa77951e209131cea741e7 |
| GPT-4o | 62.2 | Not recorded | 68fa77951e209131cea741e7 |
| Qwen2-VL-72B | 63.7 | Not recorded | 68fa77951e209131cea741e7 |
| Claude-3.5-Sonnet | 64 | Not recorded | 68fa77951e209131cea741e7 |
| MathFlow | 64.2 | Not recorded | 68fa77951e209131cea741e7 |
| VLM-R1-7B† | 65.5 | Not recorded | 68fa77951e209131cea741e7 |
| InternVL2.5-78B | 67.8 | Not recorded | 68fa77951e209131cea741e7 |
| Gemini-2.5-pro | 68.9 | Not recorded | 68fa77951e209131cea741e7 |
| GPT-4V | 72 | Not recorded | 68fa77951e209131cea741e7 |
| GPT-5 | 77 | Not recorded | 68fa77951e209131cea741e7 |
| MathFlow | 77 | Not recorded | 68fa77951e209131cea741e7 |

### InfiMM-Math-7B

CoT Error: 46.1. Result ID: 69e7384d47f34ce924cd7e05.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74da1e209131cea642cc.md>)

Recorded evaluation configuration: {"Model Type":"Open-source MLLMs"}. This record does not assert independent reproduction.

### InternVL2.5-8B

CoT Error: 49.6. Result ID: 69e7384d47f34ce924cd7e06.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bab1e209131cea89c7f.md>)

[Implementation](<https://github.com/OpenGVLab/InternVL>)

Recorded evaluation configuration: {"Model Type":"Open-source MLLMs"}. This record does not assert independent reproduction.

### Qwen-VL-Max

CoT Error: 53.9. Result ID: 69e7384d47f34ce924cd7e0c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70d41e209131cea4d774.md>)

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

Recorded evaluation configuration: {"Model Type":"Closed-source MLLMs"}. This record does not assert independent reproduction.

### GPT-4o-mini

CoT Error: 59.6. Result ID: 69e7384d47f34ce924cd7e0d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

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

Recorded evaluation configuration: {"Model Type":"Closed-source MLLMs"}. This record does not assert independent reproduction.

### Qwen2.5-VL-7B

CoT Error: 62. Result ID: 69e7384d47f34ce924cd7e07.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

Recorded evaluation configuration: {"Model Type":"Open-source MLLMs"}. This record does not assert independent reproduction.

### GPT-4o

CoT Error: 62.2. Result ID: 69e7384d47f34ce924cd7e0f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

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

Recorded evaluation configuration: {"Model Type":"Closed-source MLLMs"}. This record does not assert independent reproduction.

### Qwen2-VL-72B

CoT Error: 63.7. Result ID: 69e7384d47f34ce924cd7e09.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

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

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

Recorded evaluation configuration: {"Model Type":"Open-source MLLMs"}. This record does not assert independent reproduction.

### Claude-3.5-Sonnet

CoT Error: 64. Result ID: 69e7384d47f34ce924cd7e0e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

Recorded evaluation configuration: {"Model Type":"Closed-source MLLMs"}. This record does not assert independent reproduction.

### MathFlow

CoT Error: 64.2. Result ID: 69e7384d47f34ce924cd7e0b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

Recorded evaluation configuration: {"Model Type":"Open-source MLLMs","Backbone":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction.

### VLM-R1-7B†

CoT Error: 65.5. Result ID: 69e7384d47f34ce924cd7e08.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

Recorded evaluation configuration: {"Model Type":"Open-source MLLMs"}. This record does not assert independent reproduction.

### InternVL2.5-78B

CoT Error: 67.8. Result ID: 69e7384d47f34ce924cd7e0a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bab1e209131cea89c7f.md>)

[Implementation](<https://github.com/OpenGVLab/InternVL>)

Recorded evaluation configuration: {"Model Type":"Open-source MLLMs"}. This record does not assert independent reproduction.

### Gemini-2.5-pro

CoT Error: 68.9. Result ID: 69e7384d47f34ce924cd7e11.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76361e209131cea6be11.md>)

Recorded evaluation configuration: {"Model Type":"Closed-source MLLMs"}. This record does not assert independent reproduction.

### GPT-4V

CoT Error: 72. Result ID: 69e7384d47f34ce924cd7e10.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

Recorded evaluation configuration: {"Model Type":"Closed-source MLLMs"}. This record does not assert independent reproduction.

### GPT-5

CoT Error: 77. Result ID: 69e7384d47f34ce924cd7e12.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

Recorded evaluation configuration: {"Model Type":"Closed-source MLLMs"}. This record does not assert independent reproduction.

### MathFlow

CoT Error: 77. Result ID: 69e7384d47f34ce924cd7e13.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77951e209131cea741e7.md>)

Recorded evaluation configuration: {"Model Type":"Closed-source MLLMs","Backbone":"GPT-5"}. This record does not assert independent reproduction.

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