Overview
Public inference endpoints exposed by Aramm Gateway.
Aramm's public inference docs are organized around one landing page and one page per public endpoint. The sidebar mirrors the endpoint list, while each endpoint page uses a split layout with request and response details on the left and code and payload examples on the right.
API Reference
The pages in this section follow a reusable pattern:
Overviewis the landing page.- Each endpoint gets its own page and sidebar item.
- The right rail contains
cURL,Python, andTypeScripttabs. - The lower right block shows the response example with copy support.
Endpoints
GET /v1/models
POST /v1/chat/completions
POST /v1/messages
POST /v1/embeddings
POST /v1/audio/speech
POST /v1/audio/transcriptions
POST /v1/images/generations
What this section includes
- Model catalog lookup
- OpenAI-compatible chat and embeddings
- Anthropic-compatible messages
- Image generation and audio endpoints
What stays out
- Health, readiness, and metrics
- Guardrail test utilities
- Internal team-only surfaces
Page layout
- Left column: auth, request fields, response fields, errors, and notes.
- Right column: a stacked example rail with a tabbed code sample above a copyable response snippet.
- Mobile: the right column drops below the main content.