▲ 232 today15.1k starsPythonHN ×1
FreeToken lets your gaming PC run 290B-parameter AI models
What is FreeToken?
FreeToken is an edge-native serving engine that runs frontier-scale mixture-of-experts models on a single desktop or laptop. It streams weights, caches experts, and balances CPU-GPU work so a consumer RTX 30-series card can serve 290-billion-parameter models at interactive speed.
Who it is for
AI developers and hobbyists who want to experiment with massive open-weight models locally can skip cloud rentals and run them on existing hardware.