Introduction
What Caedral is — OpenAI-compatible AI API with subscription pools, Caedral-operated products, and Notre on eligible chat.
On this page
Caedral is an OpenAI-compatible AI API. One key covers chat, Caedral-operated Embed / Rerank / Voice, and frontier catalog models. Subscription plans include monthly usage pools. Notre Runtime is included on every plan and applied on eligible POST /v1/chat/completions when the platform runs Notre. When included quota is exhausted, enable on-demand or upgrade — otherwise the API returns HTTP 402.
Platform surface
| Capability | Endpoint | Notes |
|---|---|---|
| Chat | POST /v1/chat/completions | OpenAI-compatible messages schema — live catalog of lab model IDs. Notre Runtime may apply when the platform runs Notre. |
| Embeddings | POST /v1/embeddings | Caedral E1 Small (caedral-embed-e1-small-v1, 384D) · Caedral Models pool · not Notre |
| Rerank | POST /v1/rerank | BAAI/bge-reranker-v2-m3 · External Models pool · not Notre |
| Voice | POST /v1/audio/speech | Caedral Voice (caedral-voice-1) · English WAV · Caedral Models pool · not Notre |
| Notre | POST /v1/chat/completions field `notre` | Included on every plan; applied on eligible chat when the platform runs Notre. Stripped before upstream. |
| Models | GET /v1/models | Live public catalog |
| Usage | GET /v1/usage | Plan, pools, on-demand, and HTTP 402 state |
Every request is billed through an append-only ledger and served under your plan's included pools. Model IDs are passed through exactly as they appear in the catalog — Caedral does not rebrand third-party models.
Where to start
- Quickstart — create a key, choose a plan, run curl
- n8n integration — HTTP Request node with Bearer auth
- API reference — parameters, responses, and error codes