# Mathematical Reasoning on AIME24 (Accuracy, Tokens, Token Reduction %) Source: [Mathematical Reasoning on AIME24 (Accuracy, Tokens, Token Reduction %)]() Collection index: [llms.txt]() Data updated: 2026-07-01T08:01:56.474Z Evaluation of model performance on the AIME 2024 dataset for Mathematical Reasoning, focusing on accuracy and token efficiency metrics (raw tokens and percentage reduction). 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) | AIME24 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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. - [Tokens](): Lower is better; unit: SIZE. - [Token Reduction (%)](): Lower is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Parallel Thinking | 86 | Not recorded | 6a449f12e0224bd4f49930df | | Fork-think | 86 | Not recorded | 6a449f12e0224bd4f49930df | | Fork-think | 81.8 | Not recorded | 6a449f12e0224bd4f49930df | | Parallel Thinking | 80.1 | Not recorded | 6a449f12e0224bd4f49930df | | Greedy | 80 | Not recorded | 6a449f12e0224bd4f49930df | | CoT | 80 | Not recorded | 6a449f12e0224bd4f49930df | | Parallel Thinking | 77.8 | Not recorded | 6a449f12e0224bd4f49930df | | Fork-think | 77 | Not recorded | 6a449f12e0224bd4f49930df | | CoT | 68.7 | Not recorded | 6a449f12e0224bd4f49930df | | CoT | 67.3 | Not recorded | 6a449f12e0224bd4f49930df | | Greedy | 50 | Not recorded | 6a449f12e0224bd4f49930df | | Greedy | 43.3 | Not recorded | 6a449f12e0224bd4f49930df | ### Parallel Thinking Accuracy: 86. Result ID: 6a44c97373d097d845bef7b9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DeepSeek-8B","Decoding Strategy":"Parallel Thinking","Branching factor (n)":"32","Aggregation Strategy":"majority voting"}. This record does not assert independent reproduction. ### Fork-think Accuracy: 86. Result ID: 6a44c97373d097d845bef7ba. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DeepSeek-8B","Decoding Strategy":"Fork-think","Branching factor (n)":"32","Aggregation Strategy":"majority voting","Seed path length (l0)":"2,048"}. This record does not assert independent reproduction. ### Fork-think Accuracy: 81.8. Result ID: 6a44c97373d097d845bef7b6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-8B","Decoding Strategy":"Fork-think","Branching factor (n)":"32","Aggregation Strategy":"majority voting","Seed path length (l0)":"2,048"}. This record does not assert independent reproduction. ### Parallel Thinking Accuracy: 80.1. Result ID: 6a44c97373d097d845bef7b5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-8B","Decoding Strategy":"Parallel Thinking","Branching factor (n)":"32","Aggregation Strategy":"majority voting"}. This record does not assert independent reproduction. ### Greedy Accuracy: 80. Result ID: 6a44c97373d097d845bef7b7. [Reporting paper]() Recorded evaluation configuration: {"Model":"DeepSeek-8B","Decoding Strategy":"Greedy","Aggregation Strategy":"majority voting"}. This record does not assert independent reproduction. ### CoT Accuracy: 80. Result ID: 6a44c97373d097d845bef7b8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DeepSeek-8B","Decoding Strategy":"CoT","Aggregation Strategy":"majority voting"}. This record does not assert independent reproduction. ### Parallel Thinking Accuracy: 77.8. Result ID: 6a44c97373d097d845bef7bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Phi-4-RP-14B","Decoding Strategy":"Parallel Thinking","Branching factor (n)":"32","Aggregation Strategy":"majority voting"}. This record does not assert independent reproduction. ### Fork-think Accuracy: 77. Result ID: 6a44c97373d097d845bef7be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Phi-4-RP-14B","Decoding Strategy":"Fork-think","Branching factor (n)":"32","Aggregation Strategy":"majority voting","Seed path length (l0)":"2,048"}. This record does not assert independent reproduction. ### CoT Accuracy: 68.7. Result ID: 6a44c97373d097d845bef7b4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-8B","Decoding Strategy":"CoT","Aggregation Strategy":"majority voting"}. This record does not assert independent reproduction. ### CoT Accuracy: 67.3. Result ID: 6a44c97373d097d845bef7bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Phi-4-RP-14B","Decoding Strategy":"CoT","Aggregation Strategy":"majority voting"}. This record does not assert independent reproduction. ### Greedy Accuracy: 50. Result ID: 6a44c97373d097d845bef7b3. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen3-8B","Decoding Strategy":"Greedy","Aggregation Strategy":"majority voting"}. This record does not assert independent reproduction. ### Greedy Accuracy: 43.3. Result ID: 6a44c97373d097d845bef7bb. [Reporting paper]() Recorded evaluation configuration: {"Model":"Phi-4-RP-14B","Decoding Strategy":"Greedy","Aggregation Strategy":"majority voting"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.