POST /v1/embeddings

Embeddings generation endpoint.

Inference APIs

Generate embeddings through a configured provider and return the upstream list shape with gateway headers.

POST/v1/embeddings

Request

  • Auth: tenant auth when enabled; otherwise dev mode is open
  • Content-Type: application/json
  • Max body size: 4 MiB

Request fields

FieldTypeRequiredNotes
modelstringyesAlias or direct provider string
inputany JSONyesString, array of strings, or token array depending on upstream provider
encoding_formatstringnoPassed upstream when present
dimensionsintegernoPassed upstream when present
userstringnoPassed upstream when present
metadata or arammobjectnoAttribution fields stripped before upstream dispatch

Response schema

FieldTypeRequiredNotes
objectstringyesAlways list
modelstringyesPublic alias used for the request
dataarrayyesEmbedding vectors in index order
data[].indexintegeryesInput index
data[].objectstringyesAlways embedding
data[].embeddingnumber[]yesEmbedding vector
usageobjectnoToken usage when returned by the upstream provider
usage.prompt_tokensintegernoPrompt token count
usage.total_tokensintegernoTotal token count

Errors

HTTPCodeWhen
400invalid_request_errorMalformed JSON or missing required fields
400provider_not_configuredProvider missing from config or unavailable
400invalid_request_errorProvider does not support embeddings
502provider-specificUpstream provider error normalized to the gateway envelope

Notes

  • The gateway returns the upstream JSON shape.
  • Unsupported providers return normalized errors.
  • Response headers include model, provider, latency, and billing metadata when available.