#how to get started?

5 messages · Page 1 of 1 (latest)

fiery widget
#

how to get started?

unborn kestrelBOT
#

Hi @fiery widget. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:

#

“Getting started” can mean different things with OpenAI. Here are the main options from the provided sources:

  1. Start using the OpenAI API (developers)

    • Create an API key and try models in the Playground.
    • Follow the Developer quickstart for basic code examples. [Start exploring API]

  2. Build your first agent with the Agents SDK

    • Install the SDK (npm install @openai/agents or pip install openai-agents).
    • Set OPENAI_API_KEY.
    • Define an Agent with a name, instructions, and model, then run it with a single prompt. [Agents SDK quickstart; Python agents quickstart]

  3. Use GPTs inside ChatGPT (no coding)

    • Open Explore GPTs in ChatGPT to browse and use existing GPTs.
    • Or go to the GPTs area and click Create to build your own, then test and save it. [GPTs in ChatGPT]

  4. Use Deep Research in ChatGPT

    • In ChatGPT, type /deepresearch or pick Deep research from the tools menu.
    • Optionally connect apps (e.g., Google Drive) or specify websites under Sites → Manage sites.
    • Enter your prompt, review the proposed research plan, then run and download the report. [Deep research in ChatGPT]
#
  1. Organize work with Projects in ChatGPT
    • Click New project in the sidebar, name it, and pick an icon/color to keep related files, instructions, and chats together. [Projects in ChatGPT]

If you clarify what you want to get started with (API, agents, GPTs, Deep Research, ads, etc.), I can point you to the exact step‑by‑step flow from the docs.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:

#

@fiery widget Kapa.ai is still learning and improving, please let me know how I did by reacting below.