Model Releases

Qwen-AgentWorld open-sourced to help agents learn to predict before acting

AI summary

Tongyi Lab has open-sourced Qwen-AgentWorld, the first native language world model, covering seven domains: MCP, Search, Terminal, SWE, Web, OS and Android. It is based on more than 10 million real interaction trajectories and trained through three stages: CPT, SFT and RL.

Key Takeaways

  • Tongyi Lab has open-sourced Qwen-AgentWorld,the first native language world model
  • covering seven domains: MCP,Search,Terminal,OS and Android.
  • It is based on more than 10 million real interaction trajectories and trained through three…

Quick Facts

Published
Updated
Source
公众号:通义实验室(千问)
Category
Model Releases
AI score
79
Read the original Read the full article at the original source
Why it matters

将环境建模从预训练阶段起作为显式目标,通过可控模拟实现定向行为塑造,效果超越真实环境 RL,为高风险场景的智能体开发提供低成本方案。

Related entities

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: Qwen-AgentWorld open-sourced to help agents learn to predict before acting. June 24, 2026. https://new.nat6.net/en/news/cmsdu3gxy014yrofzv08mqwvc/

View all →