PowerInfer is a new inference engine designed to run large language models more quickly on a single consumer-grade GPU, achieving speeds up to 11.69x faster than llama.cpp. Read the paper on Hugging Face.
via Hugging Face.
PowerInfer is a new inference engine designed to run large language models more quickly on a single consumer-grade GPU, achieving speeds up to 11.69x faster than llama.cpp. Read the paper on Hugging Face.
via Hugging Face.
Bon Appétit chronicles 2023's viral food moments, from all-female M&M packages and carnivore bread to Lana Del Rey's Waffle House shift and the rise of "girl dinner." via Bon Appétit.
James Pethokoukis explores in this article if AI's recent problem-solving breakthroughs, such as one by Google's DeepMind, could lead to a productivity boom. via Faster, Please!
Tyler Cowen's latest podcast episode is a 42-minute discussion with Henry Oliver on the philosopher John Stuart Mill.
via Marginal REVOLUTION.Standard Ebooks is giving away free ebooks of classic titles with custom covers! via kottke.org
Doug Fish sold his Indy Pass, a cheaper alternative to the Epic and Ikon ski passes, to entrepreneur Erik Mogensen. via Outside Online.
Stay up to date with email notifications of new posts.