-
Uptime Kuma: open-source is great
-
Moonshine.ai adds streaming to their speech model
-
Lessons from the Edge: Optimizing YOLOv8 on an Intel Mac Mini
-
Hazy Research: Intelligence per Watt study
-
Google's multimodal LLM model, good old BART and a new Tinker.
-
Google's smallest public LLM model and Meta's neural codec
-
From idea to working code and demo in one day
-
Mentoring an AI Fellow summer project
-
Downgrading macOS with Apple Configurator
-
First post here
-
Quantizing DeepSeek R1-0528-Qwen3-8B
-
Demultiplexing music