DeepWeb-Bench - Reasoning: leaderboard

The reasoning capability family of DeepWeb-Bench (2026), a deep-research benchmark that demands massive cross-source web evidence and long-horizon derivation, scored cell by cell. The score is the mean cell score (%) over the reasoning cells for agents run in Claude Code or Codex CLI with the search and browsing tools the benchmark provides; higher is better.

Metric: Mean cell score (%) over the Reasoning capability family (multi-step reasoning over the gathered evidence), DeepWeb-Bench deep-research tasks answered cell by cell with the benchmark's web search, page visit and PDF tools (native browsing disabled), runs hosted in Claude Code CLI or Codex CLI at default settings; higher is better. Source: arxiv.org. Saturation forecast: Around June 2028. 9 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-5.532.38#54
2Claude Opus 4.731.59#85
3DeepSeek V4 Pro27.94#196
4Claude Sonnet 4.626.8#173
5GLM-5.126.7#198
6DeepSeek V4 Flash26.37#267
7Qwen 3.6 Plus25.84#238
8MiniMax-M2.723.7#539
9Kimi K2.615.13#192

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=deepweb-bench-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.