Lynkr
Route AI coding requests across models while reducing token overhead.
- Stars
- 527
- Forks
- 55
- Updated
- Updated Jul 14, 2026
Route AI coding requests across models while reducing token overhead.
Lynkr is a self-hosted LLM gateway for coding tools including Claude Code, Cursor, and Codex. After an interactive setup, users connect existing clients through a local endpoint or wrapper command, then route simple and complex work to different local or cloud models. It reduces unnecessary model traffic through tool-result compression, tool filtering, and semantic caching. The gateway requires Node.js 20+, while local configurations also need a running backend such as Ollama.
Resource types
Use cases
Platforms
Handle coding tasks through natural language in the terminal, IDE, or GitHub.
Use Gemini from the terminal to inspect code, make changes, and automate development tasks.
Runtime
Protocols & integrations
Capabilities
Audience
Public GitHub facts last synced Jul 14, 2026.
Manage multiple coding agents in parallel from one terminal dashboard.
Let AI agents control browsers through sandboxed JavaScript.