# Mathematical Reasoning on AIME 25 (Acc, Expl., Fmt) Source: [Mathematical Reasoning on AIME 25 (Acc, Expl., Fmt)]() Collection index: [llms.txt]() Data updated: 2026-04-20T06:55:35.951Z Evaluation of model performance on the AIME 25 dataset for mathematical reasoning, reporting results based on Accuracy (Acc), Explanation quality (Expl.), and Formatting correctness (Fmt). 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 25 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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. - [Explanation Quality](): Higher is better; unit: PERCENT. - [Formatting Correctness](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | LACE: + SFT & RL | 16.7 | Not recorded | 69e5b30c7d0172140ec7dfbc | | LACE: + SFT & RL | 13.3 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Independent Sampling: Standard SFT + RL | 13.3 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Independent Sampling: + Voting | 13.3 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Independent Sampling: + Voting | 10 | Not recorded | 69e5b30c7d0172140ec7dfbc | | LACE: + SFT | 10 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Independent Sampling: Standard SFT + RL | 8.3 | Not recorded | 69e5b30c7d0172140ec7dfbc | | LACE: Continuous Pre-train | 6.7 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Isolated Parallel: + SFT | 6.7 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Isolated Parallel: + SFT & RL | 6.7 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Isolated Parallel: + SFT & RL | 3.3 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Isolated Parallel: + Voting | 3.3 | Not recorded | 69e5b30c7d0172140ec7dfbc | | LACE: + SFT | 3.3 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Isolated Parallel: + SFT | 0 | Not recorded | 69e5b30c7d0172140ec7dfbc | | Isolated Parallel: + Voting | 0 | Not recorded | 69e5b30c7d0172140ec7dfbc | | LACE: Continuous Pre-train | 0 | Not recorded | 69e5b30c7d0172140ec7dfbc | ### LACE: + SFT & RL Accuracy: 16.7. Result ID: 69e5cdca47f34ce924cb739a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-4B"}. This record does not assert independent reproduction. ### LACE: + SFT & RL Accuracy: 13.3. Result ID: 69e5cdca47f34ce924cb7392. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-1.7B"}. This record does not assert independent reproduction. ### Independent Sampling: Standard SFT + RL Accuracy: 13.3. Result ID: 69e5cdca47f34ce924cb7393. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-4B"}. This record does not assert independent reproduction. ### Independent Sampling: + Voting Accuracy: 13.3. Result ID: 69e5cdca47f34ce924cb7394. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-4B"}. This record does not assert independent reproduction. ### Independent Sampling: + Voting Accuracy: 10. Result ID: 69e5cdca47f34ce924cb738c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-1.7B"}. This record does not assert independent reproduction. ### LACE: + SFT Accuracy: 10. Result ID: 69e5cdca47f34ce924cb7391. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-1.7B"}. This record does not assert independent reproduction. ### Independent Sampling: Standard SFT + RL Accuracy: 8.3. Result ID: 69e5cdca47f34ce924cb738b. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-1.7B"}. This record does not assert independent reproduction. ### LACE: Continuous Pre-train Accuracy: 6.7. Result ID: 69e5cdca47f34ce924cb7390. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-1.7B"}. This record does not assert independent reproduction. ### Isolated Parallel: + SFT Accuracy: 6.7. Result ID: 69e5cdca47f34ce924cb7395. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-4B"}. This record does not assert independent reproduction. ### Isolated Parallel: + SFT & RL Accuracy: 6.7. Result ID: 69e5cdca47f34ce924cb7396. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-4B"}. This record does not assert independent reproduction. ### Isolated Parallel: + SFT & RL Accuracy: 3.3. Result ID: 69e5cdca47f34ce924cb738e. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-1.7B"}. This record does not assert independent reproduction. ### Isolated Parallel: + Voting Accuracy: 3.3. Result ID: 69e5cdca47f34ce924cb7397. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-4B"}. This record does not assert independent reproduction. ### LACE: + SFT Accuracy: 3.3. Result ID: 69e5cdca47f34ce924cb7399. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-4B"}. This record does not assert independent reproduction. ### Isolated Parallel: + SFT Accuracy: 0. Result ID: 69e5cdca47f34ce924cb738d. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-1.7B"}. This record does not assert independent reproduction. ### Isolated Parallel: + Voting Accuracy: 0. Result ID: 69e5cdca47f34ce924cb738f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-1.7B"}. This record does not assert independent reproduction. ### LACE: Continuous Pre-train Accuracy: 0. Result ID: 69e5cdca47f34ce924cb7398. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"Qwen3-4B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.