WebHarbor released
The story so far
aiming-lab released WebHarbor, a project containing 153 executable, containerized website environments, aimed at providing stable evaluation and reinforcement learning training environments for GUI agents, supporting strong generalization across multi-step and post-authentication tasks. The suite includes 3,518 tasks, covering multi-step and post-authentication workflows, surpassing existing work. The Qwen 4B-level GUI agent trained in WebHarbor through reinforcement learning shows competitive performance in four benchmark tests compared to leading open-source GUI agents. Code and data are publicly available at https://github.com/aiming-lab/WebHarbor.
AI-written from the reports · Updated 14 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- Hugging Face Daily PapersWebHarbor: Docking Real Websites for Evolving GUI Agent Environments
WebHarbor provides 153 executable, containerized web environments for stable evaluation and reinforcement learning training of GUI agents, supporting strong generalization across multi-step and post-authentication tasks. The suite includes 3,518 tasks, covering multi-step and post-authentication workflows, surpassing existing ones. The Qwen 4B-level GUI agent trained via reinforcement learning in WebHarbor demonstrates competitive performance with leading open-source GUI agents on four benchmarks, with code and data publicly available.