Omixa API Docs

Developer documentation

Omixa API Docs

Build against Omixa with one API surface for keys, wallet billing, model routing, streaming responses, media jobs, usage reporting, and support workflows.

Documentation shortcuts
Chat and reasoning models 106 active models documented with fields, payloads, response notes, and Markdown export. Image generation and editing models 27 active models documented with fields, payloads, response notes, and Markdown export. Video generation models 13 active models documented with fields, payloads, response notes, and Markdown export. Audio and speech models 28 active models documented with fields, payloads, response notes, and Markdown export. Embedding models 10 active models documented with fields, payloads, response notes, and Markdown export. Music generation models 4 active models documented with fields, payloads, response notes, and Markdown export.

Quickstart path

Use this path when you are integrating Omixa for the first time. Create a key, add wallet balance, test in the playground, then move the same payload shape into your server code.

  • Create or copy an API key from the dashboard.
  • Choose a model from the catalog and confirm the endpoint family.
  • Send requests to the `/api/v1` base path with Bearer authentication.
  • Watch wallet holds, final cost, latency, and request status in usage screens.

Base URL and headers

All production integrations should keep keys server-side and send JSON unless an endpoint explicitly requires multipart files.

Item Value
Base URL http://omixa.cloud/api/v1
Auth header Authorization: Bearer omx_live_...
JSON header Content-Type: application/json
Streaming accept Accept: text/event-stream

First request

Chat requests follow an OpenAI-style shape. Omixa selects the active route, reserves wallet balance, executes the upstream call, then captures actual usage.

Example
curl -X POST http://omixa.cloud/api/v1/chat/completions \
  -H "Authorization: Bearer omx_live_..." \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-5","messages":[{"role":"user","content":"Hello from Omixa"}]}'

Models Catalog

The catalog is organized by workflow: chat, image, video, audio, embeddings, and music. Each model page includes request fields, response expectations, pricing context, and language examples.

  • Use `/docs/chat` for language, reasoning, tools, multimodal, and streaming.
  • Use `/docs/image` for generation, editing, image references, and binary outputs.
  • Use `/docs/video` and `/docs/music` for job-based media workflows.
  • Use `/docs/audio` and `/docs/embedding` for speech, realtime, transcription, and vectors.
Copied Markdown