创建

登录 ReadmeX

或

了解我们如何处理你的信息: 隐私政策

ReadmeX

陪你读懂 AI 动态

今天,想先了解什么?

选一份动态总结,或直接聊聊你的问题。

社区和关注动态总结只属于你自己。

资讯

SwarmWorld智能体协作在消费级GPU上复现

AI 总结

Apex Fusion合作团队使用4位量化版Qwen3.8-27B,并沿用MIT的代码和指令,在四张RTX 3090上复现了SwarmWorld实验的核心结果。此次约26小时的运行共调用模型12,800次,12项配对效应中有9项与原实验方向一致;量化模型单独工作时完成6项研究里程碑,组成智能体社会后则完成全部16项。结果仅来自单一群体规模的一次运行,尚未重跑MIT实验中更大的群体规模。

深度见解

复现实验表明,MIT报告的群体协作优势并不依赖托管的前沿模型,同时显示KV缓存精度可能显著影响自托管智能体基准测试的结果。

Apex Fusion graphic with the headline “MIT’s AI agent society research holds up on open-weight models, Apex Fusion replication shows” over a row of green and red squares
Credit: Apex Fusion
Image Credits
Credit: Apex Fusion

In August, researchers at MIT dropped a society of AI agents into a shared world with no instructions. They showed it could invent, build and maintain technology better than the same number of agents working alone. The study ran on a hosted frontier model.

A research team working with Apex Fusion has now reproduced the central experiment on four RTX 3090 gaming GPUs. Apex Fusion is a Swiss foundation building infrastructure for AI agents. It used MIT’s code and commands unchanged. The run took about 26 hours and 12,800 model calls. The paper, posted on Apex Fusion’s GitHub, and the full trace data are public.

#MIT#Apex Fusion#HAL8#SwarmWorld#Qwen3.8-27B#RTX 3090#Vector#Prime#AP3X#Agentic AI Foundation#Linux Foundation#Markus Buehler

原文约 4 分钟读完阅读原文
群聊

还没有评论,来说说你的看法。