+AOpenAI integration
OpenAI is one of the 3 model families supported by AbstractOS's multi-LLM engine. You bring your own API key, keep control of quota and cost, and the router picks GPT when it's best for the task.
What this integration does
Use your credits
Your API key = your billing = you control the cost.
Multi-LLM routes
The router picks GPT when performance or cost justifies.
New models
GPT-4o, GPT-4 Turbo, GPT-3.5 — always on the latest available.
Encrypted API key
Your key stays encrypted at rest and can be rotated at any time.
How to connect
- 1
Generate an API key
In your OpenAI account, create a dedicated secret key for AbstractOS.
- 2
Add it to Abstract
Paste it in the Prisma Studio Integrations panel.
- 3
Done
Multi-LLM starts considering OpenAI models in routing.
What we access
Transparency on the access this integration requires.
- We use the API key only to call OpenAI's API in your tasks.
- Token cost is billed straight to your OpenAI account.
- You can rotate or revoke the key at any time.
Frequently asked questions
Why API key and not OAuth?
OpenAI uses API key as standard. Billing your account requires your direct key.
Can I use only OpenAI without Claude/Gemini?
Yes. You connect only the providers you want; the router respects what is available.
Build with OpenAI on Prisma Studio
Generate your app with AI and connect OpenAI in a few clicks.
Explore Prisma Studio