Get Started
Welcome
One API for every multimodal AI provider
BasicRouter is an all-in-one API service for developers. It exposes a single OpenAI-compatible endpoint to a wide range of first-party and third-party multimodal models — chat, image generation, image editing, video generation, and beyond — under one key, one bill, one set of docs.
This site is the reference for the HTTP API. The reference covers request shape, authentication, response envelope, async tasks, error codes, and the most-used capabilities, so you can wire the platform into your product without reading source.
Where to start
What you can build
A short list of the most-used capabilities, with the highlights that matter when you are picking a model and the docs section that covers it in full.
| Capability | Highlights | Where to look |
|---|---|---|
| Chat / LLM | OpenAI-compatible Chat Completions across multiple model families | /v1/chat/completions |
| Image generation | Multiple model families, configurable resolution and aspect ratio | /ai/createImage |
| Image editing | Reference-image-guided editing and style transfer | /ai/editImage |
| Video generation | Text-to-video, image-to-video, async task + polling | /ai/createVideo |
| Models | List all available models, filter by category | /v1/models |
How the API is shaped
- Industry aliases — /v1/chat/completions and /v1/models are first-class endpoints. Anything that already speaks the OpenAI protocol can talk to the platform by changing the base URL and the API key — no SDK swap, no middleware.
- Asynchronous tasks — Image and (in particular) video generation use a POST → poll GET pattern. Submit a task, get a task_id, poll until succeeded, then read the result URL. The same pattern works across model families.
- Unified response format — All non-streaming endpoints share the same { code, message, data } envelope. The HTTP status carries transport-level errors; the envelope carries business-level errors with a stable code you can branch on.
Help
- Found a bug? Include the X-Request-Id response header when filing the issue, plus the model id and a minimal repro. It cuts the round-trip from days to hours.
- Email —admin@basic-ware.ai Documentation inquiries can be filed directly from the floating support button. Use email for account, billing, and abuse-related questions only.