Hosted inference rollout is invite-first. Abuse-resistant keys, egress controls, and model allowlists ship with enterprise workspaces.

Vocifer AI
Docs
Overview

Overview

Start here for Vocifer AI hosted inference — OpenAI-compatible HTTPS APIs and a queryable models catalog.

Vocifer AI serves OpenAI-compatible inference: one Bearer token, familiar /v1 paths, and JSON shaped like OpenAI’s Chat Completions and models list. Production traffic is typically routed to a dedicated host such as https://inf.vocifer.com (your onboarding email may assign a regional hostname).

Use the short guides below in order, or jump to what you need:

  1. Getting started — base URL, environments, and the /v1 prefix
  2. Authentication — API keys, optional org header, errors
  3. Models catalog — GET /v1/models and what the response is for
  4. Chat completions — non-streaming and streaming curl examples
  5. SDKs & clients — Python, TypeScript, Go, and third-party stacks

Exact SKUs, hosts, and headers are confirmed when your workspace is provisioned. For API access, use Request API access on the main site.

Getting started