CerberusAI Docs

One place for access, API integration, and model delivery.

Start in the access portal, integrate against the OpenAI-compatible gateway, or pull model files straight from the public CDN.

Open API reference Create account
Quick path
  1. Create an account on the access portal.
  2. Generate an API key from the dashboard.
  3. Call the Cerberus gateway or download GGUF models.
Gateway: api.cerberusai.dev
Access: access.cerberusai.dev
CDN: llm.cerberusai.dev
Reference

API Reference

OpenAI-compatible chat completions, model discovery, health checks, error codes, and direct download endpoints.

Open docs
Access

Access Portal

Sign up, verify your email, manage keys, open chat, and move into the Cerberus account surface.

Open portal
Public JSON

Model CDN

Browse the public model index, inspect exact file sizes, and link directly to hosted GGUF builds.

Browse CDN
External

HuggingFace

Jump to the public HuggingFace page for model details, metadata, and alternate discovery paths.

Open HuggingFace
Available models

Current public inventory

  • Arbiter-GL9b — 9B GLM-4 base
  • cerberus-4b-v2-abliterated — 4B refusal-ablated
  • gamma3-1b-abliterated — 1B compact, CPU-friendly

The gateway exposes the currently allowed models, and the CDN exposes downloadable model directories.

Integration notes

What to expect

  • The gateway follows the OpenAI chat shape.
  • The CDN is public and CORS-enabled.
  • Range requests are supported for large model downloads.