# Mathematical Reasoning on AIME 2025 (Seed, VeRA-E, and ∆) Source: [Mathematical Reasoning on AIME 2025 (Seed, VeRA-E, and ∆)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:04:09.781Z Evaluation of model performance on the American Invitational Mathematics Examination (AIME) 2025 dataset, focusing on the Seed, VeRA-E, and ∆ metrics. 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 2025 | | Metric | Seed Score | | 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 - [Seed Score](): Higher is better; unit: PERCENT. - [VeRA-E Score](): Higher is better; unit: PERCENT. - [Delta (∆)](): Lower is better; unit: SCALAR. ## Recorded results | Method | Seed Score | Result date | Report paper | | --- | --- | --- | --- | | GLM-4.6 | 92 | Not recorded | 6993f6229ebc499c97409b26 | | Gemini-3-Pro-Preview | 90.7 | Not recorded | 6993f6229ebc499c97409b26 | | DeepSeek-V3.2-thinking | 90 | Not recorded | 6993f6229ebc499c97409b26 | | GPT-5-high | 89.3 | Not recorded | 6993f6229ebc499c97409b26 | | GPT-5.1-high | 88.7 | Not recorded | 6993f6229ebc499c97409b26 | | DeepSeek-V3.1-thinking | 86.7 | Not recorded | 6993f6229ebc499c97409b26 | | Gemini-2.5-Pro | 85.3 | Not recorded | 6993f6229ebc499c97409b26 | | Seed-1.6-1015-high | 80.7 | Not recorded | 6993f6229ebc499c97409b26 | | Seed-1.6-Thinking-0715 | 80 | Not recorded | 6993f6229ebc499c97409b26 | | Claude-Sonnet-4.5-thinking | 78.7 | Not recorded | 6993f6229ebc499c97409b26 | | Kimi-K2-thinking | 78.7 | Not recorded | 6993f6229ebc499c97409b26 | | Seed-1.6-Lite-1015-high | 77.3 | Not recorded | 6993f6229ebc499c97409b26 | | Minimax-M2 | 76.7 | Not recorded | 6993f6229ebc499c97409b26 | | qwen3-max-0923 | 74.7 | Not recorded | 6993f6229ebc499c97409b26 | | GPT-5.1-chat-latest | 50.7 | Not recorded | 6993f6229ebc499c97409b26 | | Kimi-K2-0905 | 47.3 | Not recorded | 6993f6229ebc499c97409b26 | ### GLM-4.6 Seed Score: 92. Result ID: 699a6b615698665ec4f3d7a2. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### Gemini-3-Pro-Preview Seed Score: 90.7. Result ID: 699a6b615698665ec4f3d7aa. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### DeepSeek-V3.2-thinking Seed Score: 90. Result ID: 699a6b615698665ec4f3d7a6. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### GPT-5-high Seed Score: 89.3. Result ID: 699a6b615698665ec4f3d7ab. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### GPT-5.1-high Seed Score: 88.7. Result ID: 699a6b615698665ec4f3d7a3. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### DeepSeek-V3.1-thinking Seed Score: 86.7. Result ID: 699a6b615698665ec4f3d7b0. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### Gemini-2.5-Pro Seed Score: 85.3. Result ID: 699a6b615698665ec4f3d7a1. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### Seed-1.6-1015-high Seed Score: 80.7. Result ID: 699a6b615698665ec4f3d7a7. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### Seed-1.6-Thinking-0715 Seed Score: 80. Result ID: 699a6b615698665ec4f3d7a5. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### Claude-Sonnet-4.5-thinking Seed Score: 78.7. Result ID: 699a6b615698665ec4f3d7a4. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### Kimi-K2-thinking Seed Score: 78.7. Result ID: 699a6b615698665ec4f3d7a8. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### Seed-1.6-Lite-1015-high Seed Score: 77.3. Result ID: 699a6b615698665ec4f3d7af. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### Minimax-M2 Seed Score: 76.7. Result ID: 699a6b615698665ec4f3d7a9. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### qwen3-max-0923 Seed Score: 74.7. Result ID: 699a6b615698665ec4f3d7ad. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### GPT-5.1-chat-latest Seed Score: 50.7. Result ID: 699a6b615698665ec4f3d7ae. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. ### Kimi-K2-0905 Seed Score: 47.3. Result ID: 699a6b615698665ec4f3d7ac. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Protocol":"Avg@5 (%)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.