Tarseek API

Chat / text generation

POSThttps://api.tarseek.com/v1/chat/completions

This is the text generation API. Not every catalog model is a chat model.

Unknown JSON keys are rejected. Optional fields that a model does not support fail before a provider is called.

Minimal request

completion = client.chat.completions.create(
    model="openai/gpt-4o-mini",
    messages=[
        {"role": "system", "content": "Be brief."},
        {"role": "user", "content": "What is Tarseek?"},
    ],
)

Messages and roles

Roles: system, developer, user, assistant, tool. Content may be a string or an array of parts (text + image_url / input_image for vision).

A tool message requires tool_call_id. Only assistant messages may contain tool_calls, and only assistant tool-call messages may omit content.

Max output

Use max_completion_tokens. The deprecated max_tokens field remains accepted for compatible legacy models. Sending both is a validation error. The catalog’s supported_parameters field tells you which form a model accepts.

Sampling

temperature (0–2) and top_p (0–1] are accepted. See Parameters for the full request schema.

Stop and seed

stop is one string or 1–4 strings. seed is accepted only for models and deployments that advertise support. Unsupported optional parameters fail before routing with 400 unsupported_parameter.

Response

Every successful non-streaming response has an OpenAI-shaped id, object: "chat.completion", created, canonical model, and choices. Provider-private fields and upstream identifiers are not exposed. Normalized usage is included when the provider reports it.

The response id matches the x-request-id header. Errors include the same request ID in both the header and JSON body.