Smart Inference
Route every LLM call to the cheapest endpoint that meets your quality bar. Drop-in OpenAI-compatible.
Documentation
Everything you need to ship AI that actually works.
Guides, references and quickstarts for Smart Inference, Guard, Architect and Code. One place to find everything.
Products
Start with the product you’re using.
Guard
Policy-as-code for prompts and outputs. Author-time + CI + runtime enforcement.
Architect
Visual designer for multi-model pipelines and agent graphs. Export to code.
Code
Free, local-first control room for Claude Code, Codex CLI and Gemini CLI, on your desktop and your phone.