# Mathematical Reasoning on LogicVista Source: [Mathematical Reasoning on LogicVista]() Collection index: [llms.txt]() Data updated: 2026-06-09T07:49:22.846Z Evaluation of model performance on the LogicVista dataset for Mathematical Reasoning tasks, measured by Accuracy. 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) | LogicVista | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 54 of 54 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. - [Acc@1](): Higher is better; unit: PERCENT. - [Acc@4](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Gemini-2.5-Pro\* | 73.8 | Not recorded | 68fa7bd11e209131cea8ab14 | | GPT-5-Thinking\* | 70 | Not recorded | 68fa7bd11e209131cea8ab14 | | GPT-4o | 64.4 | Not recorded | 6926898ed6b0dd33429303b1 | | MM-Eureka-Qwen-32B | 54.6 | Not recorded | 6926898ed6b0dd33429303b1 | | MM-Eureka-Qwen-32B | 52.8 | Not recorded | 6926898ed6b0dd33429303b1 | | Gemini-2.0-Flash | 52.3 | Not recorded | 6926898ed6b0dd33429303b1 | | MM-Eureka-Qwen-7B | 51 | Not recorded | 6926898ed6b0dd33429303b1 | | VAPO-Thinker-7B | 50.9 | Not recorded | 68fa7bd11e209131cea8ab14 | | Vision-R1-7B | 49.7 | Not recorded | 68fa7bd11e209131cea8ab14 | | VL-Rethinker-7B | 49.7 | Not recorded | 6926898ed6b0dd33429303b1 | | Claude-3.5 Sonnet | 49.3 | Not recorded | 6926898ed6b0dd33429303b1 | | MM-Eureka-Qwen-7B | 49.2 | Not recorded | 6926898ed6b0dd33429303b1 | | VLAA-Thinker-7B | 48.5 | Not recorded | 68fa7bd11e209131cea8ab14 | | VL-Rethinker-7B | 46.1 | Not recorded | 6926898ed6b0dd33429303b1 | | R1-OneVision-7B | 45.6 | Not recorded | 68fa7bd11e209131cea8ab14 | | MMR1-Math-v0-7B | 44.5 | Not recorded | 6926898ed6b0dd33429303b1 | | InternVL3-8B | 43.6 | Not recorded | 6926898ed6b0dd33429303b1 | | MMR1-Math-v0-7B | 43.6 | Not recorded | 6926898ed6b0dd33429303b1 | | Kimi-VL-16B | 42.7 | Not recorded | 6926898ed6b0dd33429303b1 | | QvQ-72B-Preview | 42.7 | Not recorded | 6926898ed6b0dd33429303b1 | | Qwen2.5-VL-7B | 42.6 | Not recorded | 68fa7bd11e209131cea8ab14 | | GPT-4o-mini | 41.4 | Not recorded | 6926898ed6b0dd33429303b1 | | VAPO-Thinker-3B | 39.7 | Not recorded | 68fa7bd11e209131cea8ab14 | | InternVL2.5-8B\* | 38.3 | Not recorded | 68fa7bd11e209131cea8ab14 | | Qwen2.5-VL-7B | 35.6 | Not recorded | 6926898ed6b0dd33429303b1 | | LLaVA-OV-7B | 32 | Not recorded | 6926898ed6b0dd33429303b1 | | Qwen2.5-VL-3B | 26.6 | Not recorded | 6926898ed6b0dd33429303b1 | ### Gemini-2.5-Pro\* Accuracy: 73.8. Result ID: 69cb909547f34ce924b41b17. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Close-source","Reference Source":"OpenCompass leaderboard"}. This record does not assert independent reproduction. ### GPT-5-Thinking\* Accuracy: 70. Result ID: 69cb909547f34ce924b41b16. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Close-source","Reference Source":"OpenCompass leaderboard"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 64.4. Result ID: 69fd9f0447f34ce924e09a90. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### MM-Eureka-Qwen-32B Accuracy: 54.6. Result ID: 69fd9f0447f34ce924e09aa0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"true"}. This record does not assert independent reproduction. ### MM-Eureka-Qwen-32B Accuracy: 52.8. Result ID: 69fd9f0447f34ce924e09a9f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### Gemini-2.0-Flash Accuracy: 52.3. Result ID: 69fd9f0447f34ce924e09a8f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### MM-Eureka-Qwen-7B Accuracy: 51. Result ID: 69fd9f0447f34ce924e09a9c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"true"}. This record does not assert independent reproduction. ### VAPO-Thinker-7B Accuracy: 50.9. Result ID: 69cb909547f34ce924b41b1e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Our models","Parameter Scale":"7B"}. This record does not assert independent reproduction. ### Vision-R1-7B Accuracy: 49.7. Result ID: 69cb909547f34ce924b41b1c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Open-source","Parameter Scale":"7B"}. This record does not assert independent reproduction. ### VL-Rethinker-7B Accuracy: 49.7. Result ID: 69fd9f0447f34ce924e09a9e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"true"}. This record does not assert independent reproduction. ### Claude-3.5 Sonnet Accuracy: 49.3. Result ID: 69fd9f0447f34ce924e09a91. [Reporting paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### MM-Eureka-Qwen-7B Accuracy: 49.2. Result ID: 69fd9f0447f34ce924e09a9b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### VLAA-Thinker-7B Accuracy: 48.5. Result ID: 69cb909547f34ce924b41b1b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Open-source","Parameter Scale":"7B"}. This record does not assert independent reproduction. ### VL-Rethinker-7B Accuracy: 46.1. Result ID: 69fd9f0447f34ce924e09a9d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### R1-OneVision-7B Accuracy: 45.6. Result ID: 69cb909547f34ce924b41b1a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Open-source","Parameter Scale":"7B"}. This record does not assert independent reproduction. ### MMR1-Math-v0-7B Accuracy: 44.5. Result ID: 69fd9f0447f34ce924e09a9a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"true"}. This record does not assert independent reproduction. ### InternVL3-8B Accuracy: 43.6. Result ID: 69fd9f0447f34ce924e09a96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### MMR1-Math-v0-7B Accuracy: 43.6. Result ID: 69fd9f0447f34ce924e09a99. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### Kimi-VL-16B Accuracy: 42.7. Result ID: 69fd9f0447f34ce924e09a95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### QvQ-72B-Preview Accuracy: 42.7. Result ID: 69fd9f0447f34ce924e09a98. [Reporting paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Accuracy: 42.6. Result ID: 69cb909547f34ce924b41b18. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Open-source","Parameter Scale":"7B"}. This record does not assert independent reproduction. ### GPT-4o-mini Accuracy: 41.4. Result ID: 69fd9f0447f34ce924e09a92. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### VAPO-Thinker-3B Accuracy: 39.7. Result ID: 69cb909547f34ce924b41b1d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Our models","Parameter Scale":"3B"}. This record does not assert independent reproduction. ### InternVL2.5-8B\* Accuracy: 38.3. Result ID: 69cb909547f34ce924b41b19. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Open-source","Parameter Scale":"8B","Reference Source":"OpenCompass leaderboard"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Accuracy: 35.6. Result ID: 69fd9f0447f34ce924e09a97. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### LLaVA-OV-7B Accuracy: 32. Result ID: 69fd9f0447f34ce924e09a94. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. ### Qwen2.5-VL-3B Accuracy: 26.6. Result ID: 69fd9f0447f34ce924e09a93. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Activation Replay":"false"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.