From the course: OpenAI API: Agents
Unlock this course with a free trial
Join today to access over 24,800 courses taught by industry experts.
Introducing the OpenAI Agents SDK - OpenAI API Tutorial
From the course: OpenAI API: Agents
Introducing the OpenAI Agents SDK
- The OpenAI Agents SDK is a lightweight Python framework for building advanced multi-agent workflows. This is OpenAI's first real entry into the agent tool flow workspace, and like the other tools in this space, this SDK is model agnostic, meaning you can use this SDK to build agentic workflows using models from other vendors. Although, if you're going to build with other vendors, it might make more sense to use other tools because this SDK is specialized for OpenAI's API, and it works really well with the OpenAI API. On that note, this SDK lives separately and depends on the OpenAI API. It's sort of a hat, an abstraction layer on top of it, meaning you can work with the OpenAI API the way you've always done, but if you want to build agents, you can work with this SDK instead and it handles all the interactions with the API for you. So you're just specifying how the agents work. And the API is working under the hood. The entire SDK, including documentation and examples, is contained…
Contents
-
-
-
(Locked)
Introducing the OpenAI Agents SDK1m 50s
-
(Locked)
Core concepts: Agents, runners, handoffs, and guardrails4m 53s
-
(Locked)
Authentication1m 20s
-
(Locked)
Anatomy of a simple agent workflow3m 55s
-
(Locked)
Exploring agent examples2m 33s
-
(Locked)
Tools and capabilities3m 23s
-
(Locked)
Tracing3m 54s
-
(Locked)
-
-