# Beatra

> Agent-native multimodal creation through a Universal Skill, OAuth MCP, and REST/OpenAPI.

This is the authoritative integration map for Beatra. Start with installation for agent use, or authentication and OpenAPI for a custom REST integration.

## Start here

- [Documentation overview](https://docs.beatra.ai/en/docs): Skill, MCP, REST, and capability map
- [Quickstart](https://docs.beatra.ai/en/docs/quickstart): Install with one message and complete one secure sign-in
- [Authentication](https://docs.beatra.ai/en/docs/authentication): API keys, OAuth MCP, regions, and headers
- [Install Beatra](https://beatra.ai/install.md): Universal Skill installation instructions
- [Latest release](https://beatra.ai/install.json): Stable Universal Skill package metadata

## Capabilities

- [Capabilities overview](https://docs.beatra.ai/en/docs/capabilities): Capability matrix across MCP tools, REST endpoints, and execution modes
- [Current model lineup](https://docs.beatra.ai/en/docs/models): Live selectable model and pricing roster; runtime model discovery remains authoritative
- [Current model lineup (Markdown)](https://docs.beatra.ai/en/docs/models.md): Live model roster as agent-readable Markdown
- [Text to image](https://docs.beatra.ai/en/docs/capabilities/images/text-to-image): Generate images from text prompts
- [Image to image](https://docs.beatra.ai/en/docs/capabilities/images/image-to-image): Generate images from one or more reference images
- [Image editing](https://docs.beatra.ai/en/docs/capabilities/images/image-edit): Apply bounded instructions while preserving a base image
- [Text to video](https://docs.beatra.ai/en/docs/capabilities/videos/text-to-video): Generate video from a text prompt
- [Image to video](https://docs.beatra.ai/en/docs/capabilities/videos/image-to-video): Animate a source image
- [Reference to video](https://docs.beatra.ai/en/docs/capabilities/videos/reference-to-video): Generate video from ordered image or video references
- [Frames to video](https://docs.beatra.ai/en/docs/capabilities/videos/frames-to-video): Generate motion between first and last frames
- [Video editing](https://docs.beatra.ai/en/docs/capabilities/videos/video-edit): Apply instructions and reference-guided changes to video
- [Video extension](https://docs.beatra.ai/en/docs/capabilities/videos/video-extend): Continue a source video beyond its endpoint
- [Text to speech](https://docs.beatra.ai/en/docs/capabilities/speech/text-to-speech): Generate speech audio from text and a voice
- [Voice clone](https://docs.beatra.ai/en/docs/capabilities/speech/voice-clone): Create a reusable voice from an authorized sample
- [Voices](https://docs.beatra.ai/en/docs/capabilities/speech/voices): List, inspect, and delete voices
- [Text to music](https://docs.beatra.ai/en/docs/capabilities/music/text-to-music): Generate music from prompts and lyrics
- [Reference audio to music](https://docs.beatra.ai/en/docs/capabilities/music/reference-audio-to-music): Generate music from reference audio

## Mechanisms

- [Async tasks](https://docs.beatra.ai/en/docs/mechanisms/async-tasks): Task lifecycle, polling, cancellation, and response envelopes
- [Uploads](https://docs.beatra.ai/en/docs/mechanisms/uploads): Upload source media and reference it as an artifact
- [Callbacks](https://docs.beatra.ai/en/docs/mechanisms/callbacks): Signed terminal-state notifications, retries, and verification
- [Billing and usage](https://docs.beatra.ai/en/docs/mechanisms/billing): Credits, prepaid execution, 402 behavior, and automatic refunds
- [Errors and retries](https://docs.beatra.ai/en/docs/mechanisms/errors): Error envelopes, retryability, and error codes
- [Rate limits](https://docs.beatra.ai/en/docs/mechanisms/rate-limits): Limits and X-RateLimit response headers

## API reference

- [API reference](https://docs.beatra.ai/en/docs/api-reference): Browsable operation index generated from public OpenAPI
- [OpenAPI YAML](https://docs.beatra.ai/openapi.yaml): Canonical machine-readable public API contract
- [OpenAPI JSON](https://docs.beatra.ai/openapi.json): Canonical public API contract as JSON

## Agent Markdown

- [AI Markdown: text to image](https://docs.beatra.ai/llms/images-text-to-image.md): Self-contained text-to-image integration guide
- [AI Markdown: image to image](https://docs.beatra.ai/llms/images-image-to-image.md): Self-contained image-to-image integration guide
- [AI Markdown: image editing](https://docs.beatra.ai/llms/images-edit.md): Self-contained base-preserving image editing guide
- [AI Markdown: text to video](https://docs.beatra.ai/llms/videos-text-to-video.md): Self-contained text-to-video integration guide
- [AI Markdown: image to video](https://docs.beatra.ai/llms/videos-image-to-video.md): Self-contained image-to-video integration guide
- [AI Markdown: reference to video](https://docs.beatra.ai/llms/videos-reference-to-video.md): Self-contained reference-to-video integration guide
- [AI Markdown: frames to video](https://docs.beatra.ai/llms/videos-frames-to-video.md): Self-contained frames-to-video integration guide
- [AI Markdown: video editing](https://docs.beatra.ai/llms/videos-video-edit.md): Self-contained instruction-based video editing guide
- [AI Markdown: video extension](https://docs.beatra.ai/llms/videos-video-extend.md): Self-contained video continuation guide
- [AI Markdown: text to speech](https://docs.beatra.ai/llms/audio-text-to-speech.md): Self-contained text-to-speech integration guide
- [AI Markdown: voice clone](https://docs.beatra.ai/llms/voices-clone.md): Self-contained voice-cloning integration guide
- [AI Markdown: voices](https://docs.beatra.ai/llms/voices-management.md): Self-contained voice-management integration guide
- [AI Markdown: text to music](https://docs.beatra.ai/llms/music-text-to-music.md): Self-contained text-to-music integration guide
- [AI Markdown: reference audio to music](https://docs.beatra.ai/llms/music-reference-audio-to-music.md): Self-contained reference-audio-to-music integration guide

## Conventions

- REST operations are rooted at `https://api.beatra.ai/v1`; use API keys only in authenticated requests and never publish them.
- Prefer `/llms/*.md` guides for self-contained coding-assistant context. Every docs page also has `/md/{en|zh|ja}/docs/<path>.md`.
- Async providers converge through tasks; poll task state and use signed callbacks when documented. Billing, artifacts, and refunds follow the terminal task state.
- The live model catalog is authoritative for availability and price. Public OpenAPI is authoritative for request and response schemas.

## Support

- Email: support@beatra.ai