LARRI: open-source CLI that rents a GPU and tunnels open models to a local endpoint

gethackteam · x · 2026-09-09

LARRI is a simple open-source CLI that rents a GPU, spins up any open-weights model on it (Qwen, GLM, Devstral, etc.), and tunnels it to a fixed local endpoint—with commands to spin instances up and down. It solves the pain of maintaining your own inference infrastructure while still running local models.

Related event: LARRI: Open-Source CLI Spins Up Open Models on Rented GPUs with One Command(2 posts)→

Original post →

More from coding & agent

coding & agent channel →