# Mathematical Reasoning on AIME 2024 and 2025 (test) Source: [Mathematical Reasoning on AIME 2024 and 2025 (test)]() Collection index: [llms.txt]() Data updated: 2026-03-24T15:35:24.024Z Evaluation of model performance on the combined test sets for the American Invitational Mathematics Examination (AIME) 2024 and 2025. 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) | AIME 2024 and 2025 (test) | | Metric | Overall Performance Rate | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 21 of 21 | | 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 Performance Rate](): Higher is better; unit: PERCENT. - [Average Score](): Higher is better; unit: SCALAR. - [AIME 2024 Score](): Higher is better; unit: SCALAR. - [AIME 2025 Score](): Higher is better; unit: SCALAR. - [Resource Cost](): Lower is better; unit: SCALAR. ## Recorded results | Method | Overall Performance Rate | Result date | Report paper | | --- | --- | --- | --- | | AFlow + Unified-MAS | 67.35 | Not recorded | 69c21a9d0636e5bf4afc4990 | | MAS² + Unified-MAS | 67.35 | Not recorded | 69c21a9d0636e5bf4afc4990 | | MAS² + Unified-MAS | 66.67 | Not recorded | 69c21a9d0636e5bf4afc4990 | | MAS² | 63.27 | Not recorded | 69c21a9d0636e5bf4afc4990 | | ScoreFlow + Unified-MAS | 61.22 | Not recorded | 69c21a9d0636e5bf4afc4990 | | MAS-Zero + Unified-MAS | 59.18 | Not recorded | 69c21a9d0636e5bf4afc4990 | | AFlow | 59.18 | Not recorded | 69c21a9d0636e5bf4afc4990 | | MAS-Zero | 57.14 | Not recorded | 69c21a9d0636e5bf4afc4990 | | ScoreFlow | 57.14 | Not recorded | 69c21a9d0636e5bf4afc4990 | | ScoreFlow + Unified-MAS | 57.14 | Not recorded | 69c21a9d0636e5bf4afc4990 | | Vanilla | 55.1 | Not recorded | 69c21a9d0636e5bf4afc4990 | | AFlow + Unified-MAS | 55.1 | Not recorded | 69c21a9d0636e5bf4afc4990 | | MAS-Zero + Unified-MAS | 53.06 | Not recorded | 69c21a9d0636e5bf4afc4990 | | MAS² | 51.02 | Not recorded | 69c21a9d0636e5bf4afc4990 | | MAS-Zero | 48.98 | Not recorded | 69c21a9d0636e5bf4afc4990 | | AFlow | 48.98 | Not recorded | 69c21a9d0636e5bf4afc4990 | | ScoreFlow | 44.9 | Not recorded | 69c21a9d0636e5bf4afc4990 | | Vanilla | 42.86 | Not recorded | 69c21a9d0636e5bf4afc4990 | ### AFlow + Unified-MAS Overall Performance Rate: 67.35. Result ID: 69c2af2947f34ce924abc863. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"GPT-5-Mini"}. This record does not assert independent reproduction. ### MAS² + Unified-MAS Overall Performance Rate: 67.35. Result ID: 69c2af2947f34ce924abc867. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"GPT-5-Mini"}. This record does not assert independent reproduction. ### MAS² + Unified-MAS Overall Performance Rate: 66.67. Result ID: 69c2af2947f34ce924abc870. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"DeepSeek-V3.2"}. This record does not assert independent reproduction. ### MAS² Overall Performance Rate: 63.27. Result ID: 69c2af2947f34ce924abc866. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"GPT-5-Mini"}. This record does not assert independent reproduction. ### ScoreFlow + Unified-MAS Overall Performance Rate: 61.22. Result ID: 69c2af2947f34ce924abc865. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"GPT-5-Mini"}. This record does not assert independent reproduction. ### MAS-Zero + Unified-MAS Overall Performance Rate: 59.18. Result ID: 69c2af2947f34ce924abc861. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"GPT-5-Mini"}. This record does not assert independent reproduction. ### AFlow Overall Performance Rate: 59.18. Result ID: 69c2af2947f34ce924abc862. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"GPT-5-Mini"}. This record does not assert independent reproduction. ### MAS-Zero Overall Performance Rate: 57.14. Result ID: 69c2af2947f34ce924abc860. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"GPT-5-Mini"}. This record does not assert independent reproduction. ### ScoreFlow Overall Performance Rate: 57.14. Result ID: 69c2af2947f34ce924abc864. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"GPT-5-Mini"}. This record does not assert independent reproduction. ### ScoreFlow + Unified-MAS Overall Performance Rate: 57.14. Result ID: 69c2af2947f34ce924abc86e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"DeepSeek-V3.2"}. This record does not assert independent reproduction. ### Vanilla Overall Performance Rate: 55.1. Result ID: 69c2af2947f34ce924abc85f. [Reporting paper]() Recorded evaluation configuration: {"Target LLM":"GPT-5-Mini"}. This record does not assert independent reproduction. ### AFlow + Unified-MAS Overall Performance Rate: 55.1. Result ID: 69c2af2947f34ce924abc86c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"DeepSeek-V3.2"}. This record does not assert independent reproduction. ### MAS-Zero + Unified-MAS Overall Performance Rate: 53.06. Result ID: 69c2af2947f34ce924abc86a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"DeepSeek-V3.2"}. This record does not assert independent reproduction. ### MAS² Overall Performance Rate: 51.02. Result ID: 69c2af2947f34ce924abc86f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"DeepSeek-V3.2"}. This record does not assert independent reproduction. ### MAS-Zero Overall Performance Rate: 48.98. Result ID: 69c2af2947f34ce924abc869. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"DeepSeek-V3.2"}. This record does not assert independent reproduction. ### AFlow Overall Performance Rate: 48.98. Result ID: 69c2af2947f34ce924abc86b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"DeepSeek-V3.2"}. This record does not assert independent reproduction. ### ScoreFlow Overall Performance Rate: 44.9. Result ID: 69c2af2947f34ce924abc86d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target LLM":"DeepSeek-V3.2"}. This record does not assert independent reproduction. ### Vanilla Overall Performance Rate: 42.86. Result ID: 69c2af2947f34ce924abc868. [Reporting paper]() Recorded evaluation configuration: {"Target LLM":"DeepSeek-V3.2"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.