New Engine Runs 27B AI Model in 2GB RAM on Apple Silicon
An open-source project uses on-demand weight loading to run Google's Gemma 2 27B on M-series Macs with just 2GB of RAM, trading speed for accessibility.
section
AI
Models, labs, and the frontier
OpenAI Announces GPT-5.6, Targeting Major Efficiency Gains
OpenAI's new GPT-5.6 model aims to deliver more intelligence per dollar, focusing on efficiency gains in models, inference, and complex agentic workflows.