OpenAI 已发布 Agents API 公测版：https://openai.com/index/introducing-the-agents-api/。它为开发者提供了与运行 Codex 相同的框架和基础设施。OpenAI 托管和维护该框架。开发者可以在 OpenAI 管理的沙箱、自有基础设施或合作伙伴沙箱中运行代理的计算任务。
它可以部署吗？可以。公测版对所有开发者开放。数据仅保留在美国，且不支持零数据保留。
Agents API 是基于开源 Codex 框架（https://github.com/openai/codex）构建的托管服务。OpenAI 团队表示，为工作环境扩展 Codex 和 ChatGPT 的经验展示了长期运行的代理所需的条件。他们需要一个能够管理上下文、高效使用工具、协调子代理的框架。他们还需要能够可靠运行数天的基础设施。
官方文档：https://developers.openai.com/api/docs/guides/agents-api/overview 将 API 围绕 4 个核心概念组织：
一个会话分 4 个步骤运行。你创建会话并分配任务。然后通过流式传输或 webhook 查看进度。最后，你可以继续执行新任务或调整当前操作。
OpenAI 的公告展示了一个仅用一次调用就创建的事件调查代理：
快速入门：https://developers.openai.com/api/docs/guides/agents-api/quickstart 涵盖了 API 密钥权限和 SDK 设置。
环境选择是主要的架构决策。Agents API 支持 3 种沙箱选项，也可以不使用沙箱运行。
OpenAI 在其模型发布时，会与模型同时维护并提供版本化的框架访问。
OpenAI 的运行时比较：https://developers.openai.com/api/docs/guides/agents 给出了三种选项的定位：
OpenAI 发布了这些客户报告的数据。它们由供应商提供，而非独立基准测试。
需要与我们合作推广你的 GitHub 仓库或 Hugging Face 页面或产品发布或在线研讨会等吗？请联系：https://forms.gle/wbash1wF6efRj8G58
以实践者为中心的 AI/ML 新闻和分析，每月被超过100万名开发者和研究人员阅读。
OpenAI has released the Agents API：https://openai.com/index/introducing-the-agents-api/ in public beta. It gives developers the same harness and infrastructure that run Codex. OpenAI hosts and maintains the harness. Developers run the agent’s compute in an OpenAI-managed sandbox, their own infrastructure, or a partner sandbox.
Is it deployable? Yes. It is live for all developers in public beta. Data stays US-only, and Zero Data Retention is unsupported.
The Agents API is a managed service built on the open-source Codex harness：https://github.com/openai/codex. OpenAI team states scaling Codex and ChatGPT for Work showed what long-running agents need. They need a harness that manages context, uses tools efficiently, and coordinates subagents. They also need infrastructure that keeps them running reliably for days.
The official docs：https://developers.openai.com/api/docs/guides/agents-api/overview organize the API around 4 concepts:
A session runs in 4 steps. You create it and give it a task. Then you follow progress through streaming or webhooks. Finally, you continue with a new task or steer the current turn.
OpenAI’s announcement shows an incident-investigation agent created in a single call:
The quickstart：https://developers.openai.com/api/docs/guides/agents-api/quickstart covers API key permissions and SDK setup.
Environment choice is the main architectural decision. The Agents API supports 3 sandbox options, and it can also run without a sandbox.
OpenAI maintains the harness alongside its models, with versioned access at each model launch.
OpenAI’s runtime comparison：https://developers.openai.com/api/docs/guides/agents positions the 3 options this way:
OpenAI published these customer-reported numbers. They are vendor-supplied, not independent benchmarks.
Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? Connect with us ：https://forms.gle/wbash1wF6efRj8G58
Practitioner-first AI/ML news and analysis, read by 1M+ developers and researchers every month.
扫码直达文章详情