# Vision-Centric Understanding on MuirBench Source: [Vision-Centric Understanding on MuirBench]() Collection index: [llms.txt]() Data updated: 2026-03-05T07:07:22.367Z Evaluation of vision-centric understanding capabilities using the MuirBench dataset, 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 | Vision-Centric Understanding | | Dataset (including any stated split) | MuirBench | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 | | --- | --- | --- | --- | | GPT4o | 68 | Not recorded | 68fa7bb11e209131cea89eff | | MM-Eureka-Qwen-7B | 61.1 | Not recorded | 68fa7bb11e209131cea89eff | | VisionZero-Qwen-7B (Real-World) | 59.8 | Not recorded | 68fa7bb11e209131cea89eff | | VisionZero-Qwen-7B (CLEVR) | 59.4 | Not recorded | 68fa7bb11e209131cea89eff | | VisionZero-Qwen-7B (Chart) | 58.6 | Not recorded | 68fa7bb11e209131cea89eff | | Qwen2.5-VL-7B-Instruct | 58.2 | Not recorded | 68fa7bb11e209131cea89eff | | ViGaL-Snake+Rotation | 57.9 | Not recorded | 68fa7bb11e209131cea89eff | | VLAA-Thinker-7B | 57.1 | Not recorded | 68fa7bb11e209131cea89eff | | OpenVLThinker-7B | 52.8 | Not recorded | 68fa7bb11e209131cea89eff | | R1-OneVision-7B | 46.3 | Not recorded | 68fa7bb11e209131cea89eff | ### GPT4o Accuracy: 68. Result ID: 69a92b8a6dbb7075b9520cc6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Proprietary Model"}. This record does not assert independent reproduction. ### MM-Eureka-Qwen-7B Accuracy: 61.1. Result ID: 69a92b8a6dbb7075b9520cc9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### VisionZero-Qwen-7B (Real-World) Accuracy: 59.8. Result ID: 69a92b8a6dbb7075b9520ccf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-VL-7B","Variant":"Real-World"}. This record does not assert independent reproduction. ### VisionZero-Qwen-7B (CLEVR) Accuracy: 59.4. Result ID: 69a92b8a6dbb7075b9520ccd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-VL-7B","Variant":"CLEVR"}. This record does not assert independent reproduction. ### VisionZero-Qwen-7B (Chart) Accuracy: 58.6. Result ID: 69a92b8a6dbb7075b9520cce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-VL-7B","Variant":"Chart"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B-Instruct Accuracy: 58.2. Result ID: 69a92b8a6dbb7075b9520cc7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### ViGaL-Snake+Rotation Accuracy: 57.9. Result ID: 69a92b8a6dbb7075b9520ccc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### VLAA-Thinker-7B Accuracy: 57.1. Result ID: 69a92b8a6dbb7075b9520cca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### OpenVLThinker-7B Accuracy: 52.8. Result ID: 69a92b8a6dbb7075b9520ccb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### R1-OneVision-7B Accuracy: 46.3. Result ID: 69a92b8a6dbb7075b9520cc8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.