LLM Infrastructure

Model running with REST API documentation, configuration options, and IDE and Open WebUI integration guides.

Run open AI models locally or in the cloud with private prompts and usage-based cloud pricing.

Compress tool output and route heavy work to cheaper models for AI coding agents.

Python framework for building MCP servers, clients, and interactive applications.

Open-source local proxy connecting coding agents to free, paid, subscription, and local AI models.

Local model gateway for routing coding agents across providers, models, and accounts.