Three pieces, one core experience
LlamaNexus is a Go proxy that sits in front of llama.cpp's llama-server,
adding model routing, distributed inference across machines, an Ollama-compatible API, and LAN
discovery. Two companion apps talk to it directly:
- LlamaNexus for VS Code - chat with, explain, and edit code using your local proxy, without leaving the editor.
- LlamaNexus WebUI - a standalone browser chat client with the same core experience, for when you're not in VS Code.
Everything runs on your own hardware. LlamaNexus handles the parts that make that practical - downloading models from Hugging Face with real progress reporting, spreading a big model across multiple machines over RPC when one GPU isn't enough, and overriding context size per-model without babysitting llama.cpp's KV cache allocation yourself.
Free trial, paid plans to keep going
Every account starts with a time-limited Trial covering 1 proxy install plus 1 WebUI and 1 VS Code install. A license key from your account page is required to keep chatting once the trial ends - Basic, Pro, and Enterprise plans raise those install limits (WebUI and VS Code are tracked independently, so each plan's client limit applies to both), billed monthly or annually and managed entirely from your account page.