Qwen-CUA: A Pure Vision-Based General Computer Use Agent
rohanpaul_ai · x · 2026-08-07
An overview of the Qwen-CUA model, demonstrating that strong computer agents can be trained using only standard human interfaces: screen, mouse, and keyboard.
- Core Mechanism: Uses screenshots as a universal interface without relying on underlying code or accessibility labels, acting via mouse clicks and keystrokes.
- Training Scale: Utilized nearly 100,000 virtual processor cores across about 40,000 verifiable tasks.
- Memory Management: Retains the last 20 screenshots and action history to manage lengthy workflows.
- Performance: Scored 86.2 on the OSWorld-Verified benchmark, though a gap remains for harder long-horizon tasks.
More from coding & agent
- Opus 5.5 + Three.js demo stuns with water physics, fish and blossoms — nikola_mr64990 · 2026-10-16
- Grok-built linear regression bot claims $508K PnL in 125 days on Polymarket — Roger_M_Taylor · 2026-10-16
- Karpathy's knowledge engine workflow: Claude Code turns raw notes into a self-updating wiki — nikola_mr64990 · 2026-10-16
- Sierra and Meta unveil Personal Agent Protocol with Zapier as launch partner — sujingshen · 2026-10-16
- ArcKit adds 102 AI-narrated short films for commands across 14 community overlays — mcraddock · 2026-10-16
- Personal agents live or die on chaining tasks, not single-task ability — sujingshen · 2026-10-16