Model APIs
Models and capability discovery
List models visible to your key and choose the appropriate model id, protocol and capabilities.
Updated
GET
/v1/modelsGET
/v1/models/{model}List models
GET /v1/models returns the catalog visible to the account, key and group. Use the returned id verbatim rather than deriving it from a marketing name. Availability, pricing and permissions can change; revalidate when deploying an app or switching groups.
cURL
curl --fail-with-body https://api.token4ai.cloud/v1/models -H "Authorization: Bearer $TOKEN4AI_API_KEY"Response shape (illustrative)
{"object":"list","data":[{"id":"YOUR_MODEL_ID","object":"model","owned_by":"provider"}]}Retrieve one model
GET /v1/models/{model} reads a specific model. Encode special characters in the path parameter. If the model is absent or not authorized, do not silently choose an arbitrary default; make fallback selection an explicit application decision.
Match capabilities before pricing
- Chat and reasoning: verify protocol, output limits and context capacity.
- Tools and structured output: test a minimal schema against the selected model; compatible routing does not make all model parameters identical.
- Images, audio, embeddings and reranking require the matching capability and a configured serving route. Contact support if your group has no suitable model.
- Pricing: use the console’s current group price table and actual settlement; model discovery is not a price commitment.