展示HN：Clusterflock：一个用于网络硬件的AI调度器
嗨，HN！<p>我们创建了Clusterflock，以解决在分布式环境中管理AI代理时遇到的各种问题，包括不同的显存和内存限制，以及轻松尝试新模型的需求。<p>虽然我们专注于基础设施（这是专门为网络硬件构建的），但它还配备了强大的任务运行器（或编排器），支持多会话和异步操作。<p>以下是它的主要优势：<p>硬件感知的自动下载：它会分析你的网络硬件，并自动下载最适合你特定设置的最佳模型（目前仅支持HuggingFace）。<p>紧凑打包：通过llama.cpp实现原生并行性，你可以在同一设备上运行多个较小的模型。<p>它是完全开源的。我们希望提供一种无痛的方式来部署智能集群，也希望你能觉得它有用。<p>网站：<a href="https://clusterflock.net" rel="nofollow">https://clusterflock.net</a><p>欢迎反馈。非常感谢大家的支持。
查看原文
Hi HN!<p>We built Clusterflock to solve our own headaches with managing AI agents across distributed setups, different VRAM and RAM allowances, and the need to easily try out new models.<p>While the focus on infrastructure (we built this specifically for networked hardware) it does ship with a powerful mission runner (or orchestrator), which is multi-session and asynchronous.<p>Here is what it does best:<p>Hardware-aware auto-downloading: It profiles your networked hardware and automatically pulls down the best models for your specific setup (currently only from HuggingFace).<p>Tight packing: Native parallelism via llama.cpp, you can allow it to fit multiple smaller models on same device.<p>It is fully open-source. We wanted a painless way to deploy agentic clusters, and we hope you find it useful too.<p>Website: <a href="https://clusterflock.net" rel="nofollow">https://clusterflock.net</a><p>Happy to hear feedback. Flocks very much given.