ByteDance Released UI-TARS-1.5, A Powerful Vision-language AI Agent
ByteDance has released UI-TARS-1.5, a powerful vision-language AI agent that can see, understand, and control any screen using natural language.
2025-04-22 03:00:00 - AI Revolution
Built on Qwen-VL and trained on billions of GUI screenshots, action traces, and tutorials, it outperforms GPT-4 and Claude in desktop automation, mobile control, and real-world navigation.
With advanced perception, reasoning, and a unified action space, UI-TARS-1.5 marks a major leap in AI-powered GUI automation and humanlike computer interaction.