Welcome! This blog is about NVIDIA Jetson — NVIDIA’s small AI computers: setup, tutorials, benchmarks, and running local AI models.
My approach: measure, don’t believe. Every guide comes from a real setup — no copying vendor spec sheets, but real numbers: watts, tokens/s, temperatures.
What to expect
- Jetson Orin — setup with JetPack, step-by-step guides
- JetPack news — what each release actually delivers (measured, not marketing)
- Local AI — Ollama & llama.cpp on Jetson: which models run, how fast, at what power draw
- Cluster — connecting multiple Jetsons via Turing Pi 2: build, learning curve, payoff
This blog is inspired by jetsonhacks.com — but in English, focused on local AI without a cloud requirement.