DRFlow (DRFlow-Agent): leaderboard

Metric: Step F1 (%) on the 50-task original split of DRFlow (enterprise deep-research tasks in five domains: predict a personalized workflow of action steps from scattered heterogeneous sources): the task-wise harmonic mean of step recall and precision against the reference workflow (steps matched by an LLM judge), averaged over tasks; the authors' workflow-oriented reference agent DRFlow-Agent; higher is better. Source: arxiv.org. Saturation forecast: Around June 2028. 4 models tracked.

Top models

#ModelScore
1GPT-5.263.9
2Claude Opus 4.560.7
3DeepSeek V3.255.1
4Gemini 3.1 Pro (Preview)52.05

Interactive version: theaggregate.ai/benchmark?slug=drflow-drflow-agent · How It Works · Data refreshed daily, snapshot 2026-09-29.