OpenAI-compatible API
Call POST /v1/chat/completions using an OpenAI-compatible client. Change the base URL and key — your existing request code works within the documented compatibility scope.
Features
One OpenAI-compatible surface, plus the developer tools around it: keys, model discovery, request-level visibility, and a prepaid balance.
Call POST /v1/chat/completions using an OpenAI-compatible client. Change the base URL and key — your existing request code works within the documented compatibility scope.
Browse the models currently available to your account at GET /v1/models and in the console. No hidden inventory — you see exactly what you can call.
Test the API directly in the browser with a key you supply. The key is held in memory only and never leaves your session.
Create and revoke API keys from the console. The plaintext is shown once at creation; only a digest and a safe prefix are stored.
Inspect each request by public request ID: status, input and output tokens, request cost, and latency — no wiring required.
Add prepaid funds and pay for usage that has settled. Every charge is recorded individually and shown in Balance & Billing.
Stream Chat Completions over server-sent events. Usage is recorded when the stream settles, and charges reflect delivered output.
Every request returns a public identifier you can log and look up later in the Request Explorer.
Register a machine as an Astrodyne device, report bounded hardware/runtime telemetry, and control the bundled seller runtime.
A focused set of docs: Quickstart, authentication, models, chat completions, streaming, request IDs, errors, and usage & costs.
Smart Router
The Smart Router is a separate, customer-owned surface for sending requests through your own provider keys — distinct from Astrodyne-managed API access. Your keys are used only to proxy your own traffic.
Create an account, generate a key, and explore the console. Access to send requests is invite-only while we roll out.