DeepResearchBench: leaderboard
Evaluates deep research agents on multi-step information gathering and synthesis tasks. Tests ability to conduct thorough research across complex topics.
Metric: Average Score. Source: deepresearch-bench.github.io. Saturation forecast: Around 2029. 41 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 (High) | 55.3 | #63 (Claude Opus 4.6) |
| 2 | Claude Sonnet 4.6 (High) | 54.9 | #90 (Claude Sonnet 4.6) |
| 3 | Claude Opus 4.5 (20251101) (High) | 54.8 | #74 (Claude Opus 4.5 (20251101)) |
| 4 | GPT-5.5 (High) | 54 | #31 (GPT-5.5) |
| 5 | Claude Opus 4.6 (Medium) | 53.2 | #63 (Claude Opus 4.6) |
| 6 | Claude Opus 4.6 (Low) | 51.4 | #63 (Claude Opus 4.6) |
| 7 | Claude Sonnet 4.6 (Low) | 50.4 | #90 (Claude Sonnet 4.6) |
| 8 | Claude Opus 4.8 (High) | 50.2 | #37 (Claude Opus 4.8) |
| 9 | Gemini 3 Flash (Preview) (Low) | 49.8 | #82 (Gemini 3 Flash (Preview)) |
| 10 | GPT-5.5 (Medium) | 49.6 | #31 (GPT-5.5) |
| 11 | Claude Opus 4.8 (Low) | 49.3 | #37 (Claude Opus 4.8) |
| 12 | Gemini 3 Flash | 49 | #99 |
| 13 | GPT-5.5 (Low) | 48.7 | #31 (GPT-5.5) |
| 14 | GPT-5 (Medium) | 48.6 | #98 (GPT-5) |
| 15 | Claude Opus 4.1 (20250805) | 48.3 | #132 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=deepresearchbench · How It Works · Data refreshed daily, snapshot 2026-10-20.