# Long-form research on ResearchRubrics Source: [Long-form research on ResearchRubrics]() Collection index: [llms.txt]() Data updated: 2026-05-12T17:00:06.421Z Evaluation of model performance on the Long-form research task using the ResearchRubrics dataset, measured by the primary Score metric. 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 | Long-form research | | Dataset (including any stated split) | ResearchRubrics | | Metric | Score | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | Score | Result date | Report paper | | --- | --- | --- | --- | | Gemini Deep Research | 61.5 | Not recorded | 6a02b4b3dbc836b87d824fab | | GPT-5 + Search | 60.5 | Not recorded | 6a02b4b3dbc836b87d824fab | | OpenAI Deep Research | 59.7 | Not recorded | 6a02b4b3dbc836b87d824fab | | RubricEM-8B (RL, 1400 steps) | 50.3 | Not recorded | 6a02b4b3dbc836b87d824fab | | Tongyi DeepResearch-30B-A3B | 49.5 | Not recorded | 6a02b4b3dbc836b87d824fab | | Gemini 3.1 Pro + Search | 49.1 | Not recorded | 6a02b4b3dbc836b87d824fab | | Perplexity Deep Research | 48.7 | Not recorded | 6a02b4b3dbc836b87d824fab | | DR Tulu-8B (RL, 1900 steps) | 46.4 | Not recorded | 6a02b4b3dbc836b87d824fab | | RubricEM-8B (SFT) | 42.8 | Not recorded | 6a02b4b3dbc836b87d824fab | | WebThinker QwQ-32B | 42.2 | Not recorded | 6a02b4b3dbc836b87d824fab | | WebThinker-32B-DPO | 41.9 | Not recorded | 6a02b4b3dbc836b87d824fab | | DR Tulu-8B (SFT) | 38.4 | Not recorded | 6a02b4b3dbc836b87d824fab | | Ai2 ScholarQA – Claude Sonnet | 38.1 | Not recorded | 6a02b4b3dbc836b87d824fab | | WebExplorer-8B | 33.4 | Not recorded | 6a02b4b3dbc836b87d824fab | | Qwen3-8B + Our Search | 24.5 | Not recorded | 6a02b4b3dbc836b87d824fab | | Search-R1-7B | 0 | Not recorded | 6a02b4b3dbc836b87d824fab | ### Gemini Deep Research Score: 61.5. Result ID: 6a035c7b47f34ce924e90281. [Reporting paper]() Recorded evaluation configuration: {"Category":"Closed Deep Research"}. This record does not assert independent reproduction. ### GPT-5 + Search Score: 60.5. Result ID: 6a035c7b47f34ce924e90283. [Reporting paper]() Recorded evaluation configuration: {"Category":"Closed Deep Research"}. This record does not assert independent reproduction. ### OpenAI Deep Research Score: 59.7. Result ID: 6a035c7b47f34ce924e90284. [Reporting paper]() Recorded evaluation configuration: {"Category":"Closed Deep Research"}. This record does not assert independent reproduction. ### RubricEM-8B (RL, 1400 steps) Score: 50.3. Result ID: 6a035c7b47f34ce924e9028f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"8B","Training":"RL","Steps":"1400"}. This record does not assert independent reproduction. ### Tongyi DeepResearch-30B-A3B Score: 49.5. Result ID: 6a035c7b47f34ce924e9028a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open Deep Research Models","Backbone":"30B-A3B"}. This record does not assert independent reproduction. ### Gemini 3.1 Pro + Search Score: 49.1. Result ID: 6a035c7b47f34ce924e90282. [Reporting paper]() Recorded evaluation configuration: {"Category":"Closed Deep Research"}. This record does not assert independent reproduction. ### Perplexity Deep Research Score: 48.7. Result ID: 6a035c7b47f34ce924e90280. [Reporting paper]() Recorded evaluation configuration: {"Category":"Closed Deep Research"}. This record does not assert independent reproduction. ### DR Tulu-8B (RL, 1900 steps) Score: 46.4. Result ID: 6a035c7b47f34ce924e9028c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Open Deep Research Models","Backbone":"8B","Training":"RL","Steps":"1900"}. This record does not assert independent reproduction. ### RubricEM-8B (SFT) Score: 42.8. Result ID: 6a035c7b47f34ce924e9028e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"8B","Training":"SFT"}. This record does not assert independent reproduction. ### WebThinker QwQ-32B Score: 42.2. Result ID: 6a035c7b47f34ce924e90285. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Fixed Pipeline Deep Research","Backbone":"QwQ-32B"}. This record does not assert independent reproduction. ### WebThinker-32B-DPO Score: 41.9. Result ID: 6a035c7b47f34ce924e90286. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Fixed Pipeline Deep Research","Backbone":"32B","Alignment":"DPO"}. This record does not assert independent reproduction. ### DR Tulu-8B (SFT) Score: 38.4. Result ID: 6a035c7b47f34ce924e9028b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Category":"Open Deep Research Models","Backbone":"8B","Training":"SFT"}. This record does not assert independent reproduction. ### Ai2 ScholarQA – Claude Sonnet Score: 38.1. Result ID: 6a035c7b47f34ce924e90287. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Fixed Pipeline Deep Research","Backbone":"Claude Sonnet"}. This record does not assert independent reproduction. ### WebExplorer-8B Score: 33.4. Result ID: 6a035c7b47f34ce924e90289. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open Deep Research Models","Backbone":"8B"}. This record does not assert independent reproduction. ### Qwen3-8B + Our Search Score: 24.5. Result ID: 6a035c7b47f34ce924e9028d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Search":"Our Search"}. This record does not assert independent reproduction. ### Search-R1-7B Score: 0. Result ID: 6a035c7b47f34ce924e90288. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Category":"Open Deep Research Models","Backbone":"R1-7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.