TaskEvolving Deep Research (Agent, Without Hints): leaderboard
Metric: Rubric score (%; fact-grounded pointwise checkpoints of 500 evolved deep research tasks, evidential plus analytical, judged by Qwen3.7-Max; agent mode with Google search and web scraping tools; query without hints: a short everyday question). Source: arxiv.org. Saturation forecast: Around December 2026. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Pro | 68 |
| 2 | GLM-5.2 | 68 |
| 3 | Qwen 3.7 Max | 67 |
| 4 | GPT-5.6 Terra | 67 |
| 5 | Claude Sonnet 5 | 67 |
| 6 | Gemini 3.5 Flash | 57 |
| 7 | MiniMax-M2.5 | 42 |
| 8 | Qwen 3.6 Flash | 40 |
| 9 | Qwen 3 32B | 23 |
| 10 | Qwen 3 14B | 21 |
Interactive version: theaggregate.ai/benchmark?slug=taskevolving-deep-research-agent-without-hints · How It Works · Data refreshed daily, snapshot 2026-09-26.