# General AI Assistant Reasoning on GAIA (Web) Source: [General AI Assistant Reasoning on GAIA (Web)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:46.858Z Evaluation of General AI Assistant Reasoning capabilities specifically on the web-based subset of the GAIA dataset, measured using Accuracy, Final Gain, and Best Gain. 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 | General AI Assistant Reasoning | | Dataset (including any stated split) | GAIA (Web) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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. - [Final Gain](): Higher is better; unit: SCALAR. - [Best Gain](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Claude-3.7 | 63.33 | Not recorded | 69730077b2be0ed85a78c980 | | Claude-3.7 | 62.22 | Not recorded | 69730077b2be0ed85a78c980 | | Claude-3.7 | 62.22 | Not recorded | 69730077b2be0ed85a78c980 | | Claude-3.7 | 61.11 | Not recorded | 69730077b2be0ed85a78c980 | | Claude-3.7 | 58.89 | Not recorded | 69730077b2be0ed85a78c980 | | Claude-3.7 | 51.11 | Not recorded | 69730077b2be0ed85a78c980 | | DeepVerifier-8B | 33.33 | Not recorded | 69730077b2be0ed85a78c980 | | DeepVerifier-8B | 33.33 | Not recorded | 69730077b2be0ed85a78c980 | | GPT-4.1 | 32.22 | Not recorded | 69730077b2be0ed85a78c980 | | GPT-4.1 | 32.22 | Not recorded | 69730077b2be0ed85a78c980 | | DeepVerifier-8B | 32.22 | Not recorded | 69730077b2be0ed85a78c980 | | GPT-4.1 | 31.11 | Not recorded | 69730077b2be0ed85a78c980 | | GPT-4.1 | 31.11 | Not recorded | 69730077b2be0ed85a78c980 | | GPT-4.1 | 31.11 | Not recorded | 69730077b2be0ed85a78c980 | | DeepVerifier-8B | 31.11 | Not recorded | 69730077b2be0ed85a78c980 | | DeepVerifier-8B | 31.11 | Not recorded | 69730077b2be0ed85a78c980 | | GPT-4.1 | 28.89 | Not recorded | 69730077b2be0ed85a78c980 | | DeepVerifier-8B | 26.67 | Not recorded | 69730077b2be0ed85a78c980 | ### Claude-3.7 Accuracy: 63.33. Result ID: 699624025698665ec464a2dc. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Claude-3.7-Sonnet","Feedback Rounds":4}. This record does not assert independent reproduction. ### Claude-3.7 Accuracy: 62.22. Result ID: 699624025698665ec464a2dd. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Claude-3.7-Sonnet","Feedback Rounds":6}. This record does not assert independent reproduction. ### Claude-3.7 Accuracy: 62.22. Result ID: 699624025698665ec464a2df. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Claude-3.7-Sonnet","Feedback Rounds":10}. This record does not assert independent reproduction. ### Claude-3.7 Accuracy: 61.11. Result ID: 699624025698665ec464a2de. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Claude-3.7-Sonnet","Feedback Rounds":8}. This record does not assert independent reproduction. ### Claude-3.7 Accuracy: 58.89. Result ID: 699624025698665ec464a2db. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Claude-3.7-Sonnet","Feedback Rounds":2}. This record does not assert independent reproduction. ### Claude-3.7 Accuracy: 51.11. Result ID: 699624025698665ec464a2da. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Claude-3.7-Sonnet","Feedback Rounds":0}. This record does not assert independent reproduction. ### DeepVerifier-8B Accuracy: 33.33. Result ID: 699624025698665ec464a2ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3-8B-Base","Feedback Rounds":8}. This record does not assert independent reproduction. ### DeepVerifier-8B Accuracy: 33.33. Result ID: 699624025698665ec464a2eb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3-8B-Base","Feedback Rounds":10}. This record does not assert independent reproduction. ### GPT-4.1 Accuracy: 32.22. Result ID: 699624025698665ec464a2e1. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"GPT-4.1","Feedback Rounds":2}. This record does not assert independent reproduction. ### GPT-4.1 Accuracy: 32.22. Result ID: 699624025698665ec464a2e3. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"GPT-4.1","Feedback Rounds":6}. This record does not assert independent reproduction. ### DeepVerifier-8B Accuracy: 32.22. Result ID: 699624025698665ec464a2e9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3-8B-Base","Feedback Rounds":6}. This record does not assert independent reproduction. ### GPT-4.1 Accuracy: 31.11. Result ID: 699624025698665ec464a2e2. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"GPT-4.1","Feedback Rounds":4}. This record does not assert independent reproduction. ### GPT-4.1 Accuracy: 31.11. Result ID: 699624025698665ec464a2e4. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"GPT-4.1","Feedback Rounds":8}. This record does not assert independent reproduction. ### GPT-4.1 Accuracy: 31.11. Result ID: 699624025698665ec464a2e5. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"GPT-4.1","Feedback Rounds":10}. This record does not assert independent reproduction. ### DeepVerifier-8B Accuracy: 31.11. Result ID: 699624025698665ec464a2e7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3-8B-Base","Feedback Rounds":2}. This record does not assert independent reproduction. ### DeepVerifier-8B Accuracy: 31.11. Result ID: 699624025698665ec464a2e8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3-8B-Base","Feedback Rounds":4}. This record does not assert independent reproduction. ### GPT-4.1 Accuracy: 28.89. Result ID: 699624025698665ec464a2e0. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"GPT-4.1","Feedback Rounds":0}. This record does not assert independent reproduction. ### DeepVerifier-8B Accuracy: 26.67. Result ID: 699624025698665ec464a2e6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3-8B-Base","Feedback Rounds":0}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.