# Visual Question Answering on V\*Bench (REKEY renewal) Source: [Visual Question Answering on V\*Bench (REKEY renewal)]() Collection index: [llms.txt]() Data updated: 2026-06-23T07:34:07.624Z Evaluation of Visual Question Answering performance on the V\*Bench dataset, specifically the (REKEY renewal) version, reporting mean accuracy and robustness metrics like original drop and comparison against relabeling. 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 | Visual Question Answering | | Dataset (including any stated split) | V\*Bench (REKEY renewal) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 8 of 8 | | 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. - [Accuracy Std Dev](): Lower is better; unit: SCALAR. - [Original Drop](): Lower is better; unit: PERCENT. - [Gain vs Relabel](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Seed-2.0-Lite | 79.9 | Not recorded | 6a3a130a206542251d7c6f3d | | GPT-5.5 | 77.1 | Not recorded | 6a3a130a206542251d7c6f3d | | Qwen3.6-Plus | 75.7 | Not recorded | 6a3a130a206542251d7c6f3d | | Gemini 3.1 Pro | 75 | Not recorded | 6a3a130a206542251d7c6f3d | | MiMo v2.5 | 73.6 | Not recorded | 6a3a130a206542251d7c6f3d | | Kimi K2.6 | 72.6 | Not recorded | 6a3a130a206542251d7c6f3d | | Claude Opus 4.7 | 67.2 | Not recorded | 6a3a130a206542251d7c6f3d | | Llama 4 Maverick | 56.9 | Not recorded | 6a3a130a206542251d7c6f3d | ### Seed-2.0-Lite Accuracy: 79.9. Result ID: 6a3a35befbc373592c340b53. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-5.5 Accuracy: 77.1. Result ID: 6a3a35befbc373592c340b50. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen3.6-Plus Accuracy: 75.7. Result ID: 6a3a35befbc373592c340b4f. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Gemini 3.1 Pro Accuracy: 75. Result ID: 6a3a35befbc373592c340b52. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MiMo v2.5 Accuracy: 73.6. Result ID: 6a3a35befbc373592c340b55. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Kimi K2.6 Accuracy: 72.6. Result ID: 6a3a35befbc373592c340b54. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Claude Opus 4.7 Accuracy: 67.2. Result ID: 6a3a35befbc373592c340b51. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Llama 4 Maverick Accuracy: 56.9. Result ID: 6a3a35befbc373592c340b56. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.