Flash API

Flash API guide

Approved customers use one HTTPS base endpoint and one stable product, Flash, with the bearer key delivered during direct onboarding. Build fast, low-latency integrations over an API that keeps prompt and response content out of application, billing, and log records.

Base endpointhttps://api.teralor.com/v1

Authentication

Send the assigned key as an HTTP bearer credential in the Authorization header. Treat it as a secret. Do not put it in URLs, source control, browser code, screenshots, or support messages.

Authorization: Bearer TERALO...

Public surface

Supported routes

GET/v1/models

List Flash

Returns Flash as the sole callable product with its current capabilities and limits. Treat this response as authoritative for access.

POST/v1/chat/completions

Create a Flash chat completion

Accepts authenticated non-streaming and streaming Flash chat requests.

Flash capabilities

Reasoning effort

The Flash record returned by GET /v1/models includes capabilities.reasoning when reasoning controls are available. Use only the listed reasoning_effort values; supported values, defaults, and disable support follow the active qualified implementation. Unsupported values are rejected before dispatch.

Request example

curl https://api.teralor.com/v1/chat/completions \
  -H "Authorization: Bearer TERALO..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Flash",
    "messages": [
      {"role": "user", "content": "Summarize this text."}
    ],
    "max_tokens": 512,
    "stream": false
  }'

Set stream to true for a streaming response. Clients should handle interruptions and errors without automatically replaying a request whose dispatch status is uncertain.