Embeddings


POST /v1/embeddings
Content-Type: application/json
Authorization: Bearer sk-steinkauz-live-...

The embeddings endpoint follows the common OpenAI Embeddings API request and response shape. It is useful for clients such as retrieval pipelines and n8n flows when your configured provider exposes an embedding model.


curl -sS "$STEINKAUZ_BASE_URL/v1/embeddings" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $STEINKAUZ_API_KEY" \
  -d '{
    "model": "vercel-ai-gateway/openai/text-embedding-3-small",
    "input": ["First document", "Second document"]
  }'

Supported fields are:

model
input as one non-empty string or an array of non-empty strings
dimensions when the upstream model/provider supports it
encoding_format as float (default) or base64
user

Token-ID arrays are not accepted because their tokenization is provider-specific.

Model discovery

GET /v1/models deliberately over-lists rather than treating optional catalog metadata as a per-model allowlist. A missing modelType does not hide a model. The provider adapter must support embeddings; after that, the upstream provider decides whether the selected model is usable and returns a clean runtime error when it is not.

Embedding-capable adapters include Vercel AI Gateway, OpenAI, Azure OpenAI, OpenAI-compatible providers, Google Generative AI, Google Vertex AI, and Mistral. Anthropic’s adapter does not expose embeddings.

Embedding requests use the same API-key authentication, D×E routing decision, Budgets, usage-attempt ledger, provider-cost metadata, and Usage & Activity records as other Platform API inference.