Overview
What beatra is, what you can build today, and what's coming.
beatra is an AI generation API. Today you can build text chat with streaming responses, and you can start integrating Preview music generation, source uploads, and async task tracking.
What works today: chat is Available; models, music, single-shot uploads, and task lookup are Preview. Image, video, speech, voices, and customer callbacks remain Planned.
Where each capability stands
| Capability | Status | What to do |
|---|---|---|
| Text & chat | Available | Build against /v1/chat/completions. |
| Streaming chat | Available | Set stream: true for Server-Sent Events. |
| Models catalog | Preview | Use for discovery; keep a fallback. |
| Music generation | Preview | Create music tasks and poll /v1/tasks/{task_id}. |
| Uploads & async tasks | Preview | Use single-shot uploads and task lookup; resumable uploads are still Planned. |
| Image, video, audio, voices | Planned | Read for direction; don't call yet. |
| Customer callbacks | Planned | Design only; endpoints land later. |
What you'll work with
One base URL
https://api.beatra.ai/v1 for everything.
Standard headers
Authorization, Content-Type, Idempotency-Key, X-Request-Id.
OpenAI-compatible chat
Most OpenAI Chat Completions clients work by changing the base URL and key.
Honest roadmap
Each surface is labeled Available, Preview, or Planned.