# STEM & Puzzle on DynaMath (test) Source: [STEM & Puzzle on DynaMath (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:29.254Z Evaluation of model performance on the test split of the DynaMath dataset for STEM & Puzzle tasks, measured by Accuracy. 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 | STEM & Puzzle | | Dataset (including any stated split) | DynaMath (test) | | 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. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | MMFineReason-8B | 83.4 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | Qwen3-VL 32B | 82 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | GPT5 mini | 81.4 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | MMFineReason-4B | 80.6 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | Qwen3-VL 30B-A3B | 80.1 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | Gemini-2.5 Flash | 75.9 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | MMR1 8B | 73.6 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | Qwen3-VL 8B | 73.2 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | MMFineReason-2B | 71.4 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | HoneyBee 8B | 69.4 | Not recorded | 697c3b66a8b01e76cc8da6e1 | | OMR 7B | 69.1 | Not recorded | 697c3b66a8b01e76cc8da6e1 | ### MMFineReason-8B Accuracy: 83.4. Result ID: 6997797c5698665ec4cdd081. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Source Category":"Ours","Model Parameters":"8B","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### Qwen3-VL 32B Accuracy: 82. Result ID: 6997797c5698665ec4cdd07b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Source Category":"Open-weight VLMs","Model Parameters":"32B","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### GPT5 mini Accuracy: 81.4. Result ID: 6997797c5698665ec4cdd078. [Reporting paper]() Recorded evaluation configuration: {"Model Source Category":"Closed-source VLMs","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### MMFineReason-4B Accuracy: 80.6. Result ID: 6997797c5698665ec4cdd080. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Source Category":"Ours","Model Parameters":"4B","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### Qwen3-VL 30B-A3B Accuracy: 80.1. Result ID: 6997797c5698665ec4cdd07a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Source Category":"Open-weight VLMs","Model Parameters":"30B-A3B","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### Gemini-2.5 Flash Accuracy: 75.9. Result ID: 6997797c5698665ec4cdd077. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Source Category":"Closed-source VLMs","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### MMR1 8B Accuracy: 73.6. Result ID: 6997797c5698665ec4cdd07c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Source Category":"Open-source VLMs","Model Parameters":"8B","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### Qwen3-VL 8B Accuracy: 73.2. Result ID: 6997797c5698665ec4cdd079. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Source Category":"Open-weight VLMs","Model Parameters":"8B","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### MMFineReason-2B Accuracy: 71.4. Result ID: 6997797c5698665ec4cdd07f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Source Category":"Ours","Model Parameters":"2B","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### HoneyBee 8B Accuracy: 69.4. Result ID: 6997797c5698665ec4cdd07d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Source Category":"Open-source VLMs","Model Parameters":"8B","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. ### OMR 7B Accuracy: 69.1. Result ID: 6997797c5698665ec4cdd07e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Source Category":"Open-source VLMs","Model Parameters":"7B","Evaluation Mode":"Thinking Mode"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.