智能体
earendil.com ·
@Hacker News
Pi Durable：持久化 Agent Harness 的设计与同类产品对比
文章介绍 Pi 构建的持久化（durable）agent harness，并列举 LangChain Deep Agents、Vercel Eve、OpenAI Agents API、Anthropic Managed Agents 等同类方向。作者认为其价值在于支持无人值守的长时运行与恢复监控、将 harness 与算力分离以提升安全与扩展性，以及更易实现多人协作。评论提到该项目源码约 1.5 万行，用 GPT 与 Claude 统计分别约 15 万和 25 万 token。
AI推荐理由 介绍 Pi 构建的 durable agent harness，并对比 LangChain Deep Agents、Vercel Eve、OpenAI Agents API、Anthropic Managed Agents 等同类产品，指出持久化、harness 与算力分离、多人协作等设计动因；评论补充了源码约 1.5 万行及 GPT/Claude 分词差异，对 Agent 基础设施方向有参考价值，但为博客观点而非官方发布。
74
正文
Domain: earendil.com
Points: 69 | Comments: 6
[Top Comments]
- Very cool to see Pi build a durable agent harness too. I've been building in this space for quite some time myself [0][1] and it is a super interesting place of innovation. Less hype-y that on-your-machine coding agents, but all major players are building products in this space: LangChain Deep Agents, Vercel Eve, OpenAI Agents API, Anthropic Managed Agents, etc.
The main reasons are:
1) they are "durable", i.e. easier to make long-running in an unattended way, and easier to implement recovery, monitoring, etc
2) separating the harness from the compute brings safety and scaling benefits
3) easier to make multi-player.
[0]
[1]
- It's quite an interesting place to hack on, because it's both a well understood problem but with so many wrinkles to it. I can't count how many earlier designs we chewed through before we ended up with the final one and I would not be surprised if we learn even more about it.
- > The entire source code, without tests, is about 15,000 lines, which comes out to about 150,000 tokens with GPT and about 250,000 with Claude.
Woah, that big of a difference when it comes to token counting?