阿里新开源了35B参数的MoE模型，只激活3B，256K超长上下文，配合AgentWorldBench，研究智能体世界建模的赶紧试试。
阿里Qwen团队开源了Qwen-AgentWorld-35B-A3B模型，采用MoE架构，总参数量35B，每次推理激活3B参数，支持256K上下文长度。同时发布了AgentWorldBench基准，用于评估智能体的世界建模能力。该模型在多个现实环境模拟任务上表现优于同等规模模型。相关论文已发布于arXiv，代码和模型权重在GitHub和Hugging Face上开放。
We open-source Qwen-AgentWorld-35B-A3B (MoE, 35B/3B active, 256K context) and AgentWorldBench. Two ...
We open-source Qwen-AgentWorld-35B-A3B (MoE, 35B/3B active, 256K context) and AgentWorldBench. Two routes, one roadmap: 🔬 Build the simulator — scalable, controllable, surpassing real environments 🧠 Internalize world modeling — predict before you act Qwen-AgentWorld is our attempt to investigate how language world modeling can further expand the boundaries of general agent capabilities. Go build on it 🏃🏃♂️ 📑 Pa arxiv.org/abs/2606.24597 RKq71 📖 qwen.ai/blog?id=qwen-a… Kyhsx2 💻 G github.com/QwenLM/Qwen-Ag… vb1UZCn 🤗 Huggi huggingface.co/collections/Qw… 3QBL1TM5 🧩 Mod modelscope.cn/collections/Qw… BnGYgMWWI 💬 9 🔄 14 ❤️ 80 👀 7665 📊 26 ⚡