Understand the layer between your code and the model.
Independent, practical guides to local runtimes, hosted gateways, routers and protocols. Learn what a tool does, how requests flow, where your data goes and what to check before installing it.
Your code or agent
Claude Code, an SDK, an API or another developer tool creates the request.
Optional routing layer
A local or hosted gateway can normalize, route, retry or observe the request.
Model and upstream
The selected runtime or provider receives the request and returns a result.
Find an AI tool quickly
Search by tool name, workflow or concern. Start with the guide, then follow the official source before installing.
9Router
A local routing layer for coding assistants and AI tools that can normalize requests, track quotas and apply configured fallback routes.
Open guide →Self-hosted AI gatewayOmniRoute
A local-first gateway project that exposes one compatible endpoint and handles provider connections, routing and fallback for coding workflows.
Open guide →Local model runtimeOllama
Run supported language models locally and expose a local API that can be used by scripts, applications and compatible developer tools.
Open guide →Hosted model gatewayOpenRouter
A hosted, OpenAI-compatible gateway for accessing a changing catalog of models through one API surface.
Open guide →Self-hosted LLM gatewayLiteLLM
A gateway pattern for presenting multiple model providers behind a compatible API with routing, aliases, retries and operational controls.
Open guide →Open protocolModel Context Protocol
A protocol for connecting AI applications with external data sources, prompts and tools through a standardized interface.
Open guide →No matching AI tools. Try a broader search.
Want the concepts behind these tools?
Learn AI fundamentals, prompt engineering, AI-assisted development, routing and MCP in the learning library.