Tweet ghim

Ludion — drop your AI inference bill toward $0.
A per-request load balancer between the user's GPU and your cloud: each request runs on-device (WebGPU) when it can, your server when it can't.
github.com/Ludion-ai/Ludi…
English
Ludion
34 posts

@LudionAI
A router that decides per-request whether AI runs on the user's device or your server. Measured, not guessed. Open source. https://t.co/jE6X3wQuAT



