# General Reasoning Summary on Aggregate (GSM8K, MATH500, Minerva Math, Olympiad Bench, AIME24, AIME25, GPQA) Source: [General Reasoning Summary on Aggregate (GSM8K, MATH500, Minerva Math, Olympiad Bench, AIME24, AIME25, GPQA)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:04.816Z Evaluation of model performance on general reasoning capabilities summarized across a specific aggregate suite including GSM8K, MATH500, Minerva Math, Olympiad Bench, AIME24, AIME25, and GPQA, focusing on accuracy and response length characteristics. 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 | General Reasoning Summary | | Dataset (including any stated split) | Aggregate (GSM8K, MATH500, Minerva Math, Olympiad Bench, AIME24, AIME25, GPQA) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 11 of 11 | | 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. - [Avg Response Length](): Higher is better; unit: SCALAR. - [Avg Accuracy Change (%)](): Higher is better; unit: PERCENT. - [Length Reduction Rate (%)](): Lower is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Qwen3-4B-Thinking | 79.3 | Not recorded | 695f39e453fdd5c361743d2e | | RPAM | 75.9 | Not recorded | 695f39e453fdd5c361743d2e | | Ada-R1 | 75.5 | Not recorded | 695f39e453fdd5c361743d2e | | CoD | 74.7 | Not recorded | 695f39e453fdd5c361743d2e | | Task Arithmetic | 74.6 | Not recorded | 695f39e453fdd5c361743d2e | | TIES Merging | 74 | Not recorded | 695f39e453fdd5c361743d2e | | Average Merging | 73.5 | Not recorded | 695f39e453fdd5c361743d2e | | AIM | 73.1 | Not recorded | 695f39e453fdd5c361743d2e | | ACM | 71.3 | Not recorded | 695f39e453fdd5c361743d2e | | DARE-Linear | 70.5 | Not recorded | 695f39e453fdd5c361743d2e | | Qwen3-4B-Instruct | 66.4 | Not recorded | 695f39e453fdd5c361743d2e | ### Qwen3-4B-Thinking Accuracy: 79.3. Result ID: 69969e065698665ec4e78d02. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Base Model","Thinking Strategy":"Long-CoT"}. This record does not assert independent reproduction. ### RPAM Accuracy: 75.9. Result ID: 69969e065698665ec4e78d0c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Data-dependent Merging"}. This record does not assert independent reproduction. ### Ada-R1 Accuracy: 75.5. Result ID: 69969e065698665ec4e78d05. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Training-based"}. This record does not assert independent reproduction. ### CoD Accuracy: 74.7. Result ID: 69969e065698665ec4e78d04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Prompt-guided"}. This record does not assert independent reproduction. ### Task Arithmetic Accuracy: 74.6. Result ID: 69969e065698665ec4e78d07. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Data-free Merging"}. This record does not assert independent reproduction. ### TIES Merging Accuracy: 74. Result ID: 69969e065698665ec4e78d08. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Data-free Merging"}. This record does not assert independent reproduction. ### Average Merging Accuracy: 73.5. Result ID: 69969e065698665ec4e78d06. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Method Category":"Data-free Merging"}. This record does not assert independent reproduction. ### AIM Accuracy: 73.1. Result ID: 69969e065698665ec4e78d0a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Data-dependent Merging"}. This record does not assert independent reproduction. ### ACM Accuracy: 71.3. Result ID: 69969e065698665ec4e78d0b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Data-dependent Merging"}. This record does not assert independent reproduction. ### DARE-Linear Accuracy: 70.5. Result ID: 69969e065698665ec4e78d09. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Method Category":"Data-free Merging"}. This record does not assert independent reproduction. ### Qwen3-4B-Instruct Accuracy: 66.4. Result ID: 69969e065698665ec4e78d03. [Reporting paper]() Recorded evaluation configuration: {"Method Category":"Base Model","Thinking Strategy":"Short-CoT"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.