# Video Spatial Understanding on VSI-Bench 500 QA 1.0 Source: [Video Spatial Understanding on VSI-Bench 500 QA 1.0]() Collection index: [llms.txt]() Data updated: 2026-06-11T09:56:17.026Z Evaluation of model performance on the VSI-Bench 500 QA version 1.0 dataset, focusing on Video Spatial Understanding capabilities, 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 | Video Spatial Understanding | | Dataset (including any stated split) | VSI-Bench 500 QA 1.0 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 25 of 25 | | 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 | | --- | --- | --- | --- | | CoCoSI | 49.7 | Not recorded | 6a28f32090966ade919c1637 | | CoCoSI | 49.7 | Not recorded | 6a28f32090966ade919c1637 | | Single-Agent Baseline | 44.9 | Not recorded | 6a28f32090966ade919c1637 | | Single-Agent Baseline | 44.9 | Not recorded | 6a28f32090966ade919c1637 | | TraveLER | 44.1 | Not recorded | 6a28f32090966ade919c1637 | | TraveLER | 44.1 | Not recorded | 6a28f32090966ade919c1637 | | VCA | 43.6 | Not recorded | 6a28f32090966ade919c1637 | | VCA | 43.6 | Not recorded | 6a28f32090966ade919c1637 | | VideoAgent | 43.3 | Not recorded | 6a28f32090966ade919c1637 | | VideoAgent | 43.3 | Not recorded | 6a28f32090966ade919c1637 | | CoCoSI | 43 | Not recorded | 6a28f32090966ade919c1637 | | CoCoSI | 41.1 | Not recorded | 6a28f32090966ade919c1637 | | CoCoSI | 41.1 | Not recorded | 6a28f32090966ade919c1637 | | Single-Agent Baseline | 39.7 | Not recorded | 6a28f32090966ade919c1637 | | TraveLER | 39.3 | Not recorded | 6a28f32090966ade919c1637 | | VCA | 38.9 | Not recorded | 6a28f32090966ade919c1637 | | VideoAgent | 38.4 | Not recorded | 6a28f32090966ade919c1637 | | Single-Agent Baseline | 36.6 | Not recorded | 6a28f32090966ade919c1637 | | Single-Agent Baseline | 36.6 | Not recorded | 6a28f32090966ade919c1637 | | TraveLER | 36.2 | Not recorded | 6a28f32090966ade919c1637 | | TraveLER | 36.2 | Not recorded | 6a28f32090966ade919c1637 | | VCA | 35.7 | Not recorded | 6a28f32090966ade919c1637 | | VCA | 35.7 | Not recorded | 6a28f32090966ade919c1637 | | VideoAgent | 35.4 | Not recorded | 6a28f32090966ade919c1637 | | VideoAgent | 35.4 | Not recorded | 6a28f32090966ade919c1637 | ### CoCoSI Accuracy: 49.7. Result ID: 6a2a8495fbc373592c281d1b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Gemini-based","Training-free":"true","Agentic Collaboration":"true","CogMap":"true"}. This record does not assert independent reproduction. ### CoCoSI Accuracy: 49.7. Result ID: 6a2a8495fbc373592c281d3e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Gemini-2.5-Flash","Training-free":true,"Agentic Collaboration":true,"CogMap":true}. This record does not assert independent reproduction. ### Single-Agent Baseline Accuracy: 44.9. Result ID: 6a2a8495fbc373592c281d17. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Gemini-based","Training-free":"true","Agentic Collaboration":"false","CogMap":"true"}. This record does not assert independent reproduction. ### Single-Agent Baseline Accuracy: 44.9. Result ID: 6a2a8495fbc373592c281d3a. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Gemini-2.5-Flash","Training-free":true,"Agentic Collaboration":false,"CogMap":true}. This record does not assert independent reproduction. ### TraveLER Accuracy: 44.1. Result ID: 6a2a8495fbc373592c281d18. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Gemini-based","Training-free":"true","Agentic Collaboration":"true","CogMap":"false"}. This record does not assert independent reproduction. ### TraveLER Accuracy: 44.1. Result ID: 6a2a8495fbc373592c281d3b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Gemini-2.5-Flash","Training-free":true,"Agentic Collaboration":true,"CogMap":false}. This record does not assert independent reproduction. ### VCA Accuracy: 43.6. Result ID: 6a2a8495fbc373592c281d1a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Gemini-based","Training-free":"true","Agentic Collaboration":"false","CogMap":"false"}. This record does not assert independent reproduction. ### VCA Accuracy: 43.6. Result ID: 6a2a8495fbc373592c281d3d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Gemini-2.5-Flash","Training-free":true,"Agentic Collaboration":false,"CogMap":false}. This record does not assert independent reproduction. ### VideoAgent Accuracy: 43.3. Result ID: 6a2a8495fbc373592c281d19. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Gemini-based","Training-free":"true","Agentic Collaboration":"false","CogMap":"false"}. This record does not assert independent reproduction. ### VideoAgent Accuracy: 43.3. Result ID: 6a2a8495fbc373592c281d3c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Gemini-2.5-Flash","Training-free":true,"Agentic Collaboration":false,"CogMap":false}. This record does not assert independent reproduction. ### CoCoSI Accuracy: 43. Result ID: 6a2a8495fbc373592c281d16. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"GPT-based","Training-free":"true","Agentic Collaboration":"true","CogMap":"true"}. This record does not assert independent reproduction. ### CoCoSI Accuracy: 41.1. Result ID: 6a2a8495fbc373592c281d20. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen-VL-based","Training-free":"true","Agentic Collaboration":"true","CogMap":"true"}. This record does not assert independent reproduction. ### CoCoSI Accuracy: 41.1. Result ID: 6a2a8495fbc373592c281d34. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-32B","Training-free":"true","Agentic Collaboration":"true","CogMap":"true"}. This record does not assert independent reproduction. ### Single-Agent Baseline Accuracy: 39.7. Result ID: 6a2a8495fbc373592c281d12. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"GPT-based","Training-free":"true","Agentic Collaboration":"false","CogMap":"true"}. This record does not assert independent reproduction. ### TraveLER Accuracy: 39.3. Result ID: 6a2a8495fbc373592c281d13. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"GPT-based","Training-free":"true","Agentic Collaboration":"true","CogMap":"false"}. This record does not assert independent reproduction. ### VCA Accuracy: 38.9. Result ID: 6a2a8495fbc373592c281d15. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"GPT-based","Training-free":"true","Agentic Collaboration":"false","CogMap":"false"}. This record does not assert independent reproduction. ### VideoAgent Accuracy: 38.4. Result ID: 6a2a8495fbc373592c281d14. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"GPT-based","Training-free":"true","Agentic Collaboration":"false","CogMap":"false"}. This record does not assert independent reproduction. ### Single-Agent Baseline Accuracy: 36.6. Result ID: 6a2a8495fbc373592c281d1c. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen-VL-based","Training-free":"true","Agentic Collaboration":"false","CogMap":"true"}. This record does not assert independent reproduction. ### Single-Agent Baseline Accuracy: 36.6. Result ID: 6a2a8495fbc373592c281d30. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-32B","Training-free":"true","Agentic Collaboration":"false","CogMap":"true"}. This record does not assert independent reproduction. ### TraveLER Accuracy: 36.2. Result ID: 6a2a8495fbc373592c281d1d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen-VL-based","Training-free":"true","Agentic Collaboration":"true","CogMap":"false"}. This record does not assert independent reproduction. ### TraveLER Accuracy: 36.2. Result ID: 6a2a8495fbc373592c281d31. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-32B","Training-free":"true","Agentic Collaboration":"true","CogMap":"false"}. This record does not assert independent reproduction. ### VCA Accuracy: 35.7. Result ID: 6a2a8495fbc373592c281d1f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen-VL-based","Training-free":"true","Agentic Collaboration":"false","CogMap":"false"}. This record does not assert independent reproduction. ### VCA Accuracy: 35.7. Result ID: 6a2a8495fbc373592c281d33. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-32B","Training-free":"true","Agentic Collaboration":"false","CogMap":"false"}. This record does not assert independent reproduction. ### VideoAgent Accuracy: 35.4. Result ID: 6a2a8495fbc373592c281d1e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen-VL-based","Training-free":"true","Agentic Collaboration":"false","CogMap":"false"}. This record does not assert independent reproduction. ### VideoAgent Accuracy: 35.4. Result ID: 6a2a8495fbc373592c281d32. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-32B","Training-free":"true","Agentic Collaboration":"false","CogMap":"false"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.