ByteDance Open Source Multimodal AI Agent—UI-TARS-1.5
2025-04-23 07:46:33

ByteDance has open-sourced the latest 1.5 version of its multimodal AI Agent UI-TARS. Compared with the previous generation, version 1.5 performed very well in benchmarks such as computer use, browser use, and mobile phone use. In terms of computer use, the OSworld test score was 42.5, higher than Open AI CUA's 36.4, Claude 3.7's 28, and the previous highest level of 38.1 (200 steps); Windows Agent Arena (50 steps) scored 42.1, far exceeding the previous 29.8.
Email Subscription
Newsletters and emails are now available! Delivered on time, every weekday, to keep you up to date with North American business news.
ASIA TECH WIRE

Grasp technology trends

Download