# Question Answering on CORTEXBENCH Bridge Source: [Question Answering on CORTEXBENCH Bridge]() Collection index: [llms.txt]() Data updated: 2026-06-30T10:10:48.855Z Evaluation of Question Answering performance on the CORTEXBENCH Bridge dataset, typically measured using 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 | Question Answering | | Dataset (including any stated split) | CORTEXBENCH Bridge | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 32 of 32 | | 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 | | --- | --- | --- | --- | | GPT-5.4 | 96.1 | Not recorded | 6a434d96ba73165dbb743c43 | | DeepSeek-V3.2 | 94.8 | Not recorded | 6a434d96ba73165dbb743c43 | | Claude Opus 4.6 | 91.9 | Not recorded | 6a434d96ba73165dbb743c43 | | DeepSeek-V4-Flash | 90.2 | Not recorded | 6a434d96ba73165dbb743c43 | | Claude Sonnet 4.6 | 84 | Not recorded | 6a434d96ba73165dbb743c43 | | Gemini 3.1 Pro | 82.5 | Not recorded | 6a434d96ba73165dbb743c43 | | Gemini 3 Pro | 80.1 | Not recorded | 6a434d96ba73165dbb743c43 | | GPT-4o | 77.5 | Not recorded | 6a434d96ba73165dbb743c43 | | Gemini 3.1 Pro | 65.7 | Not recorded | 6a434d96ba73165dbb743c43 | | Gemini 3 Pro | 63 | Not recorded | 6a434d96ba73165dbb743c43 | | Claude Opus 4.6 | 62.5 | Not recorded | 6a434d96ba73165dbb743c43 | | GPT-5.4 | 60.9 | Not recorded | 6a434d96ba73165dbb743c43 | | Claude Sonnet 4.6 | 59.1 | Not recorded | 6a434d96ba73165dbb743c43 | | GPT-5.4 | 58.4 | Not recorded | 6a434d96ba73165dbb743c43 | | DeepSeek-V4-Flash | 54.1 | Not recorded | 6a434d96ba73165dbb743c43 | | Gemini 3.1 Pro | 52.5 | Not recorded | 6a434d96ba73165dbb743c43 | | Claude Sonnet 4.6 | 52.5 | Not recorded | 6a434d96ba73165dbb743c43 | | DeepSeek-V3.2 | 52.2 | Not recorded | 6a434d96ba73165dbb743c43 | | Claude Opus 4.6 | 50.9 | Not recorded | 6a434d96ba73165dbb743c43 | | GPT-5.4 | 47.2 | Not recorded | 6a434d96ba73165dbb743c43 | | DeepSeek-V4-Flash | 44 | Not recorded | 6a434d96ba73165dbb743c43 | | DeepSeek-V3.2 | 41.3 | Not recorded | 6a434d96ba73165dbb743c43 | | Gemini 3 Pro | 38.1 | Not recorded | 6a434d96ba73165dbb743c43 | | Claude Opus 4.6 | 35.3 | Not recorded | 6a434d96ba73165dbb743c43 | | Claude Sonnet 4.6 | 34.7 | Not recorded | 6a434d96ba73165dbb743c43 | | GPT-4o | 33.8 | Not recorded | 6a434d96ba73165dbb743c43 | | DeepSeek-V4-Flash | 30.9 | Not recorded | 6a434d96ba73165dbb743c43 | | Gemini 3 Pro | 30.9 | Not recorded | 6a434d96ba73165dbb743c43 | | DeepSeek-V3.2 | 30.3 | Not recorded | 6a434d96ba73165dbb743c43 | | Gemini 3.1 Pro | 28.1 | Not recorded | 6a434d96ba73165dbb743c43 | | GPT-4o | 26.9 | Not recorded | 6a434d96ba73165dbb743c43 | | GPT-4o | 26.6 | Not recorded | 6a434d96ba73165dbb743c43 | ### GPT-5.4 Accuracy: 96.1. Result ID: 6a43960dfbc373592c965220. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"ORACLE"}. This record does not assert independent reproduction. ### DeepSeek-V3.2 Accuracy: 94.8. Result ID: 6a43960dfbc373592c965204. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"ORACLE"}. This record does not assert independent reproduction. ### Claude Opus 4.6 Accuracy: 91.9. Result ID: 6a43960dfbc373592c965218. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"ORACLE"}. This record does not assert independent reproduction. ### DeepSeek-V4-Flash Accuracy: 90.2. Result ID: 6a43960dfbc373592c965208. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"ORACLE"}. This record does not assert independent reproduction. ### Claude Sonnet 4.6 Accuracy: 84. Result ID: 6a43960dfbc373592c965214. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"ORACLE"}. This record does not assert independent reproduction. ### Gemini 3.1 Pro Accuracy: 82.5. Result ID: 6a43960dfbc373592c965210. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"ORACLE"}. This record does not assert independent reproduction. ### Gemini 3 Pro Accuracy: 80.1. Result ID: 6a43960dfbc373592c96520c. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"ORACLE"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 77.5. Result ID: 6a43960dfbc373592c96521c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"ORACLE"}. This record does not assert independent reproduction. ### Gemini 3.1 Pro Accuracy: 65.7. Result ID: 6a43960dfbc373592c96520e. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"1-SRCH"}. This record does not assert independent reproduction. ### Gemini 3 Pro Accuracy: 63. Result ID: 6a43960dfbc373592c96520b. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"M-SRCH"}. This record does not assert independent reproduction. ### Claude Opus 4.6 Accuracy: 62.5. Result ID: 6a43960dfbc373592c965217. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"M-SRCH"}. This record does not assert independent reproduction. ### GPT-5.4 Accuracy: 60.9. Result ID: 6a43960dfbc373592c96521f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"M-SRCH"}. This record does not assert independent reproduction. ### Claude Sonnet 4.6 Accuracy: 59.1. Result ID: 6a43960dfbc373592c965213. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"M-SRCH"}. This record does not assert independent reproduction. ### GPT-5.4 Accuracy: 58.4. Result ID: 6a43960dfbc373592c96521e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"1-SRCH"}. This record does not assert independent reproduction. ### DeepSeek-V4-Flash Accuracy: 54.1. Result ID: 6a43960dfbc373592c965207. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"M-SRCH"}. This record does not assert independent reproduction. ### Gemini 3.1 Pro Accuracy: 52.5. Result ID: 6a43960dfbc373592c96520f. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"M-SRCH"}. This record does not assert independent reproduction. ### Claude Sonnet 4.6 Accuracy: 52.5. Result ID: 6a43960dfbc373592c965212. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"1-SRCH"}. This record does not assert independent reproduction. ### DeepSeek-V3.2 Accuracy: 52.2. Result ID: 6a43960dfbc373592c965203. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"M-SRCH"}. This record does not assert independent reproduction. ### Claude Opus 4.6 Accuracy: 50.9. Result ID: 6a43960dfbc373592c965216. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"1-SRCH"}. This record does not assert independent reproduction. ### GPT-5.4 Accuracy: 47.2. Result ID: 6a43960dfbc373592c96521d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"CLOSED"}. This record does not assert independent reproduction. ### DeepSeek-V4-Flash Accuracy: 44. Result ID: 6a43960dfbc373592c965206. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"1-SRCH"}. This record does not assert independent reproduction. ### DeepSeek-V3.2 Accuracy: 41.3. Result ID: 6a43960dfbc373592c965202. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"1-SRCH"}. This record does not assert independent reproduction. ### Gemini 3 Pro Accuracy: 38.1. Result ID: 6a43960dfbc373592c96520a. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"1-SRCH"}. This record does not assert independent reproduction. ### Claude Opus 4.6 Accuracy: 35.3. Result ID: 6a43960dfbc373592c965215. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"CLOSED"}. This record does not assert independent reproduction. ### Claude Sonnet 4.6 Accuracy: 34.7. Result ID: 6a43960dfbc373592c965211. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"CLOSED"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 33.8. Result ID: 6a43960dfbc373592c96521b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"M-SRCH"}. This record does not assert independent reproduction. ### DeepSeek-V4-Flash Accuracy: 30.9. Result ID: 6a43960dfbc373592c965205. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"CLOSED"}. This record does not assert independent reproduction. ### Gemini 3 Pro Accuracy: 30.9. Result ID: 6a43960dfbc373592c965209. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"CLOSED"}. This record does not assert independent reproduction. ### DeepSeek-V3.2 Accuracy: 30.3. Result ID: 6a43960dfbc373592c965201. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"CLOSED"}. This record does not assert independent reproduction. ### Gemini 3.1 Pro Accuracy: 28.1. Result ID: 6a43960dfbc373592c96520d. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"CLOSED"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 26.9. Result ID: 6a43960dfbc373592c96521a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"1-SRCH"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 26.6. Result ID: 6a43960dfbc373592c965219. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"CLOSED"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.