DeepWeb-Bench: leaderboard
Deep research, in which an agent searches the open web, collects evidence, and derives an answer through extended reasoning, is a prominent use case for frontier language models. Frontier deep research products score high on existing benchmarks, making it difficult to distinguish their capabilities from current evaluation data alone. We introduce DeepWeb-Bench, a deep research benchmark that is su
Metric: Overall score (self-reported). Source: benchmarklist.com. Saturation forecast: Around August 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 33.37 | #31 |
| 2 | Claude Opus 4.7 | 31.84 | #49 |
| 3 | DeepSeek V4 Pro | 28.68 | #102 |
| 4 | GLM-5.1 | 28.18 | #97 |
| 5 | Claude Sonnet 4.6 | 27.97 | #90 |
| 6 | DeepSeek V4 Flash | 27.73 | #137 |
| 7 | Qwen 3.6 Plus | 26.54 | #120 |
| 8 | MiniMax-M2.7 | 24.06 | #258 |
| 9 | Kimi K2.6 | 16.79 | #104 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=deepweb-bench · How It Works · Data refreshed daily, snapshot 2026-10-20.