Simon Du Reveals SOTA Deep Research via Model Training

Simon Shaolei Du shared his team's SOTA approach to deep research, which involves training models natively rather than relying on external scaffolding, with a system capable of running up to 150 sub-agents.

2026-07-30 ~ 2026-07-30 · 2 related posts

• Achieving SOTA Deep Research: Training Models Over Scaffolding — SimonShaoleiDu · 2026-07-30
• Inside SOTA Deep Research: Native Model Training and 150 Sub-Agents — SimonShaoleiDu · 2026-07-30