umbrelOS Demo: Plug in an eGPU and Local AI Jumps from 7 to 70 tokens/sec
JosephJacks_ · x · 2026-10-02
umbrel showcased running private local AI at home on umbrelOS with no drivers, no configuration, and no command line. In a demo on an AI PC and a mini PC, plugging in an eGPU instantly boosted inference from 7 to 70 tokens/sec.
More from Infra
- huggingface_hub v2.1.0 ships: 17x faster downloads, job retries, rerun and port exposure — huggingface · 2026-10-02
- Infra engineers compare notes on when self-hosting LLMs beats paying for APIs — One_Mention_5385 · 2026-10-02
- Gemini 4 isn't even out yet — but Google's TPU advantage is being underappreciated — haider1 · 2026-10-02
- parakeet_cuda: local Parakeet transcription on a 2GB VRAM GTX 750 Ti, ~20x realtime — mgostIH · 2026-10-02
- Sentdex gets hands on Asus ET900N G3, the GB300-based DGX Station — Sentdex · 2026-10-02
- Bittensor's iota productizes decentralized training with pipeline parallelism — markjeffrey · 2026-10-02