# Multimodal Mathematical Reasoning on MathVista (GEO, ALG, GPS, MWP, ALL) Source: [Multimodal Mathematical Reasoning on MathVista (GEO, ALG, GPS, MWP, ALL)]() Collection index: [llms.txt]() Data updated: 2026-03-04T07:33:56.286Z Evaluation of model performance on the MathVista dataset for Multimodal Mathematical Reasoning, reporting scores across GEO, ALG, GPS, MWP, and ALL categories. 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 | Multimodal Mathematical Reasoning | | Dataset (including any stated split) | MathVista | | Metric | Overall Accuracy (ALL) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 (ALL)](): Higher is better; unit: PERCENT. - [GEO Accuracy](): Higher is better; unit: PERCENT. - [ALG Accuracy](): Higher is better; unit: PERCENT. - [GPS Accuracy](): Higher is better; unit: PERCENT. - [MWP Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Overall Accuracy (ALL) | Result date | Report paper | | --- | --- | --- | --- | | Vision-R1-72B\* | 78.2 | Not recorded | 68fa77591e209131cea72979 | | Vision-R1-32B\* | 76.4 | Not recorded | 68fa77591e209131cea72979 | | OpenAI O1 | 73.9 | Not recorded | 68fa77591e209131cea72979 | | Qwen2.5-VL-72B | 73.5 | Not recorded | 68fa77591e209131cea72979 | | Vision-R1-7B | 73.5 | Not recorded | 68fa77591e209131cea72979 | | Qwen2.5-VL-32B | 72.9 | Not recorded | 68fa77591e209131cea72979 | | InternVL2.5-78B | 71.9 | Not recorded | 68fa77591e209131cea72979 | | Qwen2.5-VL-7B | 68.1 | Not recorded | 68fa77591e209131cea72979 | | Claude-3.5 Sonnet | 67.7 | Not recorded | 68fa77591e209131cea72979 | | GPT-4o | 63.8 | Not recorded | 68fa77591e209131cea72979 | | Mulberry-7B | 63.1 | Not recorded | 68fa77591e209131cea72979 | | URSA-8B | 59.8 | Not recorded | 68fa77591e209131cea72979 | | LLaVA-CoT-11B | 54.8 | Not recorded | 68fa77591e209131cea72979 | | Multimath-7B | 50 | Not recorded | 68fa77591e209131cea72979 | | GPT-4V | 49.9 | Not recorded | 68fa77591e209131cea72979 | | Math-PUMA-Qwen2-7B | 47.9 | Not recorded | 68fa77591e209131cea72979 | | Math-LLaVA-13B | 46.6 | Not recorded | 68fa77591e209131cea72979 | ### Vision-R1-72B\* Overall Accuracy (ALL): 78.2. Result ID: 69a7dfa56dbb7075b94c63b0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"72B","additional data during RL training":"true"}. This record does not assert independent reproduction. ### Vision-R1-32B\* Overall Accuracy (ALL): 76.4. Result ID: 69a7dfa56dbb7075b94c63af. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"32B","additional data during RL training":"true"}. This record does not assert independent reproduction. ### OpenAI O1 Overall Accuracy (ALL): 73.9. Result ID: 69a7dfa56dbb7075b94c63a0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"-"}. This record does not assert independent reproduction. ### Qwen2.5-VL-72B Overall Accuracy (ALL): 73.5. Result ID: 69a7dfa56dbb7075b94c63a6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"72B"}. This record does not assert independent reproduction. ### Vision-R1-7B Overall Accuracy (ALL): 73.5. Result ID: 69a7dfa56dbb7075b94c63ae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"7B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-32B Overall Accuracy (ALL): 72.9. Result ID: 69a7dfa56dbb7075b94c63a5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"32B"}. This record does not assert independent reproduction. ### InternVL2.5-78B Overall Accuracy (ALL): 71.9. Result ID: 69a7dfa56dbb7075b94c63a7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Params.":"78B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Overall Accuracy (ALL): 68.1. Result ID: 69a7dfa56dbb7075b94c63a4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"7B"}. This record does not assert independent reproduction. ### Claude-3.5 Sonnet Overall Accuracy (ALL): 67.7. Result ID: 69a7dfa56dbb7075b94c63a3. [Reporting paper]() Recorded evaluation configuration: {"Params.":"-"}. This record does not assert independent reproduction. ### GPT-4o Overall Accuracy (ALL): 63.8. Result ID: 69a7dfa56dbb7075b94c63a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"-"}. This record does not assert independent reproduction. ### Mulberry-7B Overall Accuracy (ALL): 63.1. Result ID: 69a7dfa56dbb7075b94c63ad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"7B"}. This record does not assert independent reproduction. ### URSA-8B Overall Accuracy (ALL): 59.8. Result ID: 69a7dfa56dbb7075b94c63ab. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"8B"}. This record does not assert independent reproduction. ### LLaVA-CoT-11B Overall Accuracy (ALL): 54.8. Result ID: 69a7dfa56dbb7075b94c63ac. [Reporting paper]() Recorded evaluation configuration: {"Params.":"11B"}. This record does not assert independent reproduction. ### Multimath-7B Overall Accuracy (ALL): 50. Result ID: 69a7dfa56dbb7075b94c63aa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"7B"}. This record does not assert independent reproduction. ### GPT-4V Overall Accuracy (ALL): 49.9. Result ID: 69a7dfa56dbb7075b94c63a2. [Reporting paper]() Recorded evaluation configuration: {"Params.":"-"}. This record does not assert independent reproduction. ### Math-PUMA-Qwen2-7B Overall Accuracy (ALL): 47.9. Result ID: 69a7dfa56dbb7075b94c63a9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"7B"}. This record does not assert independent reproduction. ### Math-LLaVA-13B Overall Accuracy (ALL): 46.6. Result ID: 69a7dfa56dbb7075b94c63a8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params.":"13B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.