DeepResearch Bench: leaderboard
Benchmark for deep research agents, evaluating long-form research reports across comprehensiveness, insight, instruction following, readability, and citation quality.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Around February 2028. 45 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 49.71 | #152 |
| 2 | dr-tulu | 45.49 | #195 |
| 3 | Doubao Deep Research | 44.34 | #182 |
| 4 | Tongyi DeepResearch 30B A3B | 40.46 | #345 |
| 5 | Claude 3.7 Sonnet | 36.63 | #249 |
| 6 | Sonar Pro | 36.19 | #269 |
| 7 | Gemini 2.5 Pro (Preview 05-06) | 31.9 | #147 |
| 8 | GPT-4o Search Preview | 30.74 | #455 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=deepresearch-bench · How It Works · Data refreshed daily, snapshot 2026-10-20.