Generate embedding vectors for one or more inputs. Only embeddings-capable models are eligible; a non-embeddings model is rejected 404 not_found_error (embeddings_not_supported).

Authentication

Authenticated with a secret key (rd_sk_…) or a client token. A client token must carry the ai.embed scope, or the request fails 403 permission_error (insufficient_scope). A publishable key (rd_pk_…) is mint-only and is rejected 403 (publishable_key_mint_only).

Request body

Example

Response

200 OK — one vector per input, in order:
The response id equals the X-Request-Id response header. Embeddings support the same idempotency, retries, limits, budgets, and error envelope as /v1/chat.