Concepts
Providers (LLM & Embedding)
A single, unified way to connect any AI provider — for both chat and embedding models.
UnderOcean uses a unified provider layer as the single integration point for every model provider — both for chat completion and for embeddings. This means agents, flows, and knowledge bases all configure "providers" the same way, regardless of which vendor is behind them.
Supported provider types
| Provider | Extra field |
|---|---|
| OpenAI | — |
| Anthropic | — |
| Google Gemini | — |
| Ollama | — |
| Azure OpenAI | Deployment name |
| AWS Bedrock | — |
| Custom (OpenAI-compatible) | Base URL |
Choosing Custom exposes a base URL field, so any self-hosted or OpenAI-compatible endpoint can be plugged in without code changes.
Security
- API keys are encrypted at rest and never stored or logged in plaintext.
- API keys are never included in any API response — the encrypted value never leaves the backend.
- Each provider configuration can be tested without exposing the underlying key to the client.