OpenAI
Connect an OpenAI API key for models, embeddings, vision, speech, and realtime voice.
Provider
openaiConnect with
API key
Capabilities
Provider siteLLM, embeddings, vision, image generation, STT, TTS, realtime voice, web search, code interpreter
platform.openai.com
OpenAI is a model provider. Use it as the LLM (and optional speech or realtime voice) on agents, and for embeddings and image tools when those capabilities are enabled.
Connecting, testing, and toggling capabilities is the same for every provider — see Third Party Services.
Get an API key on OpenAI
- Open the OpenAI Platform and sign in.
- Open API keys.
- Create a new secret key. Copy it once — OpenAI does not show it again.
- Confirm the project has billing and access to the models you need (GPT, embeddings, Whisper, TTS, or Realtime).
Use a project API key from the Platform, not a ChatGPT Plus login. Keys live under the organization or project you billed.
Connect in Vettero
Open Integrations → Third Party Services, add OpenAI, then Connect Now.
| Setting | Description |
|---|---|
| Capabilities | Turn on the features this connection should expose |
| API Key | The secret key from OpenAI |
Host extras such as web search, code interpreter, and image generation cannot be enabled on platform-managed OpenAI connections. Use your own key when you need those.
Use it
| Place | How |
|---|---|
| Agents | LLM, realtime LLM, STT, TTS |
| Knowledge bases | Embeddings when you choose OpenAI as the embedding provider on a vector store |
| Workflows | OpenAI blocks that bind this connection |