# Web Agent Search and Reasoning on xbench deepsearch Source: [Web Agent Search and Reasoning on xbench deepsearch]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:32:47.923Z Evaluation of model performance on Web Agent Search and Reasoning capabilities using the xbench deepsearch dataset, measured by Acc, F-AE, Rounds, and Token usage. 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 | Web Agent Search and Reasoning | | Dataset (including any stated split) | xbench deepsearch | | 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. - [F-AE](): Lower is better; unit: SCALAR. - [Rounds](): Lower is better; unit: SCALAR. - [Token Usage](): Lower is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | WebClipper (Hybrid) | 73.3 | Not recorded | 6992a46f9ebc499c97406ba0 | | Tongyi-DeepResearch | 71.3 | Not recorded | 6992a46f9ebc499c97406ba0 | | WebClipper (Eff) | 71.3 | Not recorded | 6992a46f9ebc499c97406ba0 | | Prompt Control | 67.6 | Not recorded | 6992a46f9ebc499c97406ba0 | | OpenAI o3 | 67 | Not recorded | 6992a46f9ebc499c97406ba0 | | Claude-4-Sonnet | 64.6 | Not recorded | 6992a46f9ebc499c97406ba0 | | Coarse Prune | 60.3 | Not recorded | 6992a46f9ebc499c97406ba0 | | Kimi-K2-Instruct-0905 | 54 | Not recorded | 6992a46f9ebc499c97406ba0 | | WebExplorer | 51.7 | Not recorded | 6992a46f9ebc499c97406ba0 | | Qwen3-235B-A22B-Instruct-2507 | 49 | Not recorded | 6992a46f9ebc499c97406ba0 | | DeepSeek-R1-671B | 42.7 | Not recorded | 6992a46f9ebc499c97406ba0 | ### WebClipper (Hybrid) Accuracy: 73.3. Result ID: 699a526e5698665ec4da0d13. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Pruning Method","Base Model":"Tongyi-DeepResearch","Variant":"Hybrid"}. This record does not assert independent reproduction. ### Tongyi-DeepResearch Accuracy: 71.3. Result ID: 699a526e5698665ec4da0d0f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-sourced Agent","Evaluation Environment":"Unified"}. This record does not assert independent reproduction. ### WebClipper (Eff) Accuracy: 71.3. Result ID: 699a526e5698665ec4da0d12. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Pruning Method","Base Model":"Tongyi-DeepResearch","Variant":"Efficiency"}. This record does not assert independent reproduction. ### Prompt Control Accuracy: 67.6. Result ID: 699a526e5698665ec4da0d10. [Reporting paper]() Recorded evaluation configuration: {"Category":"Pruning Method","Base Model":"Tongyi-DeepResearch"}. This record does not assert independent reproduction. ### OpenAI o3 Accuracy: 67. Result ID: 699a526e5698665ec4da0d09. [Reporting paper]() Recorded evaluation configuration: {"Category":"Close-sourced System"}. This record does not assert independent reproduction. ### Claude-4-Sonnet Accuracy: 64.6. Result ID: 699a526e5698665ec4da0d0a. [Reporting paper]() Recorded evaluation configuration: {"Category":"Close-sourced System"}. This record does not assert independent reproduction. ### Coarse Prune Accuracy: 60.3. Result ID: 699a526e5698665ec4da0d11. [Reporting paper]() Recorded evaluation configuration: {"Category":"Pruning Method","Base Model":"Tongyi-DeepResearch"}. This record does not assert independent reproduction. ### Kimi-K2-Instruct-0905 Accuracy: 54. Result ID: 699a526e5698665ec4da0d0b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-sourced Agent","Evaluation Environment":"Unified"}. This record does not assert independent reproduction. ### WebExplorer Accuracy: 51.7. Result ID: 699a526e5698665ec4da0d0e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-sourced Agent"}. This record does not assert independent reproduction. ### Qwen3-235B-A22B-Instruct-2507 Accuracy: 49. Result ID: 699a526e5698665ec4da0d0d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-sourced Agent","Evaluation Environment":"Unified"}. This record does not assert independent reproduction. ### DeepSeek-R1-671B Accuracy: 42.7. Result ID: 699a526e5698665ec4da0d0c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open-sourced Agent","Evaluation Environment":"Unified"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.