# Qwen introduces RecreationBench for hybrid computer-use agents

ModelScope said Qwen has introduced RecreationBench, a 250-task benchmark for Hybrid Computer-Use Agents across Ubuntu, macOS, Windows, Android, and Web. Agents must explore a running app, recreate it in code, and pass programmatic tests plus VLM-based visual evaluation.

Language: en
Time zone: America/Los_Angeles
Current time (UTC, cached as of): 2026-11-10T07:31:00.000Z
Query date (America/Los_Angeles): 2026-11-09

HTML: https://didcodexreset.com/news/e61745cd196d4afb9b057365.html

Content language: en
Localization state: sameLanguage

Source: ModelScope · Published 9/20/2026, 23:19:03

ModelScope said **Qwen has introduced RecreationBench**, a benchmark for Hybrid Computer-Use Agents with **250 application-recreation tasks** across Ubuntu, macOS, Windows, Android, and Web. The tasks span domains such as productivity, development, graphics, multimedia, and science.

Unlike GUI-only or terminal-only benchmarks, agents must explore a running reference app, recreate it in code, and pass both programmatic tests and VLM-based visual evaluation. The dataset is available at [https://modelscope.ai/datasets/Qwen/RecreationBench](https://modelscope.ai/datasets/Qwen/RecreationBench).

Tags: Qwen, RecreationBench, Computer-use agents, Benchmark, ModelScope

[View original post](https://x.com/ModelScope2022/status/2101918026719215702)