Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Qwen-AgentWorld: Language World Models for General Agents
| Date | Stars |
|---|---|
| 2026-08-05 | 929 |
| 2026-08-06 | 929 |
| 2026-08-13 | 946 |
| 2026-08-18 | 956 |
| 2026-08-19 | 957 |
| 2026-08-20 | 962 |
| 2026-08-22 | 963 |
| 2026-08-23 | 964 |
| 2026-08-24 | 965 |
| 2026-08-26 | 972 |
| 2026-08-27 | 974 |
| 2026-08-28 | 978 |
| 2026-08-30 | 980 |
| 2026-08-31 | 983 |
| 2026-09-02 | 984 |
| 2026-09-03 | 985 |
| 2026-09-04 | 987 |
| 2026-09-05 | 988 |
| 2026-09-06 | 991 |
| 2026-09-07 | 993 |
| 2026-09-08 | 996 |
| 2026-09-09 | 1000 |
| 2026-09-13 | 1003 |
| 2026-09-14 | 1003 |
| 2026-09-16 | 1006 |
| 2026-09-17 | 1008 |
| 2026-09-18 | 1009 |
| 2026-09-19 | 1013 |
| 2026-09-20 | 1015 |
| 2026-09-23 | 1014 |
| 2026-09-25 | 1015 |
| 2026-09-26 | 1014 |
| 2026-09-27 | 1015 |
| 2026-09-28 | 1017 |
| 2026-09-29 | 1018 |
| 2026-09-30 | 1018 |
| 2026-10-01 | 1019 |
| 2026-10-02 | 1020 |
| 2026-10-03 | 1022 |
| 2026-10-04 | 1024 |
| 2026-10-05 | 1025 |
| 2026-10-06 | 1024 |
| 2026-10-09 | 1028 |
| 2026-10-10 | 1027 |
| 2026-10-12 | 1027 |
| 2026-10-13 | 1027 |
| 2026-10-14 | 1028 |
| 2026-10-15 | 1027 |
| 2026-10-17 | 1026 |
| 2026-10-19 | 1028 |
| 2026-10-20 | 1029 |
| 2026-10-21 | 1030 |
| 2026-10-24 | 1032 |
| 2026-10-25 | 1033 |
| 2026-10-27 | 1033 |
| 2026-10-28 | 1034 |
| 2026-10-29 | 1034 |
| 2026-10-30 | 1036 |
| 2026-10-31 | 1036 |
| 2026-11-01 | 1037 |
| 2026-11-03 | 1038 |
| 2026-11-04 | 1040 |
Today
+2 stars today
This week
+6 stars this week
This month
+15 stars this month
Momentum
0.0
growth rate 0.58%/day
# Qwen-AgentWorld
<div style="text-align: center">
<img width="400px" src="assets/logo.png">
<p>
<a href="http://arxiv.org/abs/2606.24597">📑 Technical Report</a> |
<a href="https://qwen.ai/blog?id=qwen-agentworld">📖 Blog</a> |
<a href="https://huggingface.co/collections/Qwen/qwen-agentworld">🤗 Hugging Face</a> |
<a href="https://modelscope.cn/collections/Qwen/Qwen-AgentWorld">🤖 ModelScope</a> |
<a href="https://qwen.ai/blog?id=qwen-agentworld#interactive-demo-interactive-demo">🖥️ Demo</a>
</p>
</div>
<div align="center">
<img src="assets/group_capybaras_flat.png" alt="Qwen-AgentWorld" width="1000px"/>
</div>
Welcome to the GitHub repository of Qwen-AgentWorld. Here, you can find official information about Qwen-AgentWorld, post your questions ([Issues](https://github.com/QwenLM/Qwen-AgentWorld/issues)), and share your ideas with the community ([Discussions](https://github.com/QwenLM/Qwen-AgentWorld/discussions)).
## News
- **2026-06-24**: We release **Qwen-AgentWorld-35B-A3B** and **AgentWorldBench**. Read more on our [blog](https://qwen.ai/blog?id=qwen-agentworld) and the [technical report](http://arxiv.org/abs/2606.24597).
## Open-Source Release
We open-source Qwen-AgentWorld-35B-A3B (model weights) and AgentWorldBench (evaluation benchmark):
| Release | Description |
|---------|-------------|
| [Qwen-AgentWorld-35B-A3B](https://huggingface.co/Qwen/Qwen-AgentWorld-35B-A3B) | Language world model (MoE, 35B total / 3B active, 256K context) |
| [AgentWorldBench](https://huggingface.co/datasets/Qwen/AgentWorldBench) | Evaluation benchmark across 7 domains |
The official weights and data are released on:
- [🤗 HuggingFace](https://huggingface.co/Qwen): Download automatically via model ID, e.g., `Qwen/Qwen-AgentWorld-35B-A3B`. You can also download model files manually using `huggingface download` or `git clone`. Please follow the instructions on the model page.
- [🤖 ModelScope](https://modelscope.cn/organization/qwen): For users unable to access Hugging Face Hub. For supported frameworks, you can download from ModelScope by setting environment variables, such as `SGLANG_USE_MODELSCOPE=true` or `VLLM_USE_MODELSCOPE=true`.
## Introduction
<div align="center">
<img src="assets/teaser.png" alt="Qwen-AgentWorld Overview" width="800px"/>
</div>
<br>
**Qwen-AgentWorld** is a native language world model that simulates agentic environments via long chain-of-thought reasoning across **seven unified domains**: MCP, Search, Terminal, SWE, Android, Web, and OS. It is trained through a three-stage pipeline -- CPT injects environment knowledge, SFT activates next-state-prediction reasoning, RL sharpens simulation fidelity -- on more than 10M real-world interaction trajectories. Unlike prior approaches that treat world modeling as a post-hoc add-on, Qwen-AgentWorld is a **native world model**: environment modeling is the training objective from the CPT stage onward.
Key features:
- **Seven Unified Domains.** The first language world model to cover seven agent interaction domains within a single model.
- **Native World Model.** Environment modeling from CPT onward, not post-hoc adaptation.
- **Generalizable, Scalable & Controllable Simulator.** Zero-shot generalization to OOD environments (e.g., Claw Agent); controllable perturbations and fictional-world construction surpass real-environment training.
- **Agent Foundation Model.** LWM RL warm-up on single-turn, non-agentic trajectories transfers to multi-turn, tool-calling agentic tasks across seven benchmarks, including three entirely out-of-domain.
## Performance
<div align="center">
<img src="assets/bench_results_bar.png" alt="AgentWorldBench Results" width="800px"/>
</div>
Five-dimensional rubric mean (↑) per domain, normalized to 0--100 scale.
| Model | MCP | Search | Term. | SWE | Android | Web | OS | **Overall** |
|:------|:---:|:------:|:-----:|:---:|:-------:|:---:|:--:|:-----------:|
| GPT-5.4 | **70.10** | 37.26 | 53.69 | 66.29Excerpt of 14,579 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:f3082ed39514c1b1, llm:Repository name and description: 'Qwen-AgentWorld: Language World Models for General Agents' — indicates language/world models and agents (likely research/implementation of agent architectures using language models).
matched fp:f3082ed39514c1b1, llm:Repository name and description: 'Qwen-AgentWorld: Language World Models for General Agents' — indicates language/world models and agents (likely research/implementation of agent architectures using language models).
matched fp:f3082ed39514c1b1, llm:Repository name and description: 'Qwen-AgentWorld: Language World Models for General Agents' — indicates language/world models and agents (likely research/implementation of agent architectures using language models).
