The API contract

• A model store. The catalog lists the models we sell with their published prices, context windows, and data posture. You call a model id from that catalog.
• One inference document. Requests are the official OpenAI Responses document — the same shape as the SDKs you already use — and responses are official Responses objects with metered usage.
• One Sylphx key. Keys start with sylphx_sk_ and are created and revoked in the Sylphx console. Apps read the key from SYLPHX_API_KEY.

Inference

Catalog

Files

POST /responses HTTP/1.1
Host: api.sylphx.ai
Authorization: Bearer sylphx_sk_…
Content-Type: application/json

• Keys are shown once when created in the Sylphx console and can be revoked there at any time; a revoked key stops authenticating.
• A key sees only its own account’s stored responses, files, and usage — another account’s object id is a 404, not a leak.
• Apps read the key from the SYLPHX_API_KEY environment variable and never ship it to a browser.

{
  "code": "rate_limit_exceeded",
  "error": {
    "code": "rate_limit_exceeded",
    "message": "Client request rate limit for this key was reached.",
    "param": null,
    "type": "rate_limit_exceeded"
  },
  "summary": "…",
  "retryable": true,
  "retry_after_seconds": 3,
  "next_action": "Reduce request rate for this key, then retry after the advertised window."
}

Status codes, the full code table, retry rules, and rate-limit headers are on errors and retries.

• Responses — the primary wire and the schema in the download. Use it for new code.
• Messages — a second public encoding: POST /messages accepts the Anthropic Messages format, normalizes once into the same Responses document, and can call any catalog model.
• Chat Completions — a second public encoding too: POST /chat/completions accepts the OpenAI Chat Completions format, normalizes once into the same Responses document, and answers chat.completion objects and chat.completion.chunk streams.