TaskEvolving Deep Research (Agent, With 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 with hints: analysis dimensions and solving constraints stated). Source: arxiv.org. Saturation forecast: Around December 2026. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Terra | 86 |
| 2 | DeepSeek V4 Pro | 83 |
| 3 | Qwen 3.7 Max | 83 |
| 4 | GLM-5.2 | 80 |
| 5 | Claude Sonnet 5 | 80 |
| 6 | Gemini 3.5 Flash | 74 |
| 7 | MiniMax-M2.5 | 67 |
| 8 | Qwen 3.6 Flash | 53 |
| 9 | Qwen 3 32B | 36 |
| 10 | Qwen 3 14B | 35 |
Interactive version: theaggregate.ai/benchmark?slug=taskevolving-deep-research-agent-with-hints · How It Works · Data refreshed daily, snapshot 2026-09-26.