- Llama 3.2 introduces small vision LLMs (11B and 90B parameters) and lightweight text-only models (1B and 3B) for edge/mobile devices, with the smaller models supporting 128K token context.
- The 11B and 90B vision models are competitive with leading closed models like Claude 3 Haiku on image understanding tasks, while being open and customizable.
- Llama 3.2 comes with official Llama Stack distributions to simplify deployment across environments (cloud, on-prem, edge), including support for RAG and safety features.
- The lightweight 1B and 3B models are optimized for on-device use cases like summarization and instruction following.
Highly recommend The Man from the Future: The Visionary Life of John von Neumann by Ananyo Bhattacharya. Talks about this time plus many of his other contributions to e.g. game theory.
Did this in north Alabama back in the mid 90s for a school fundraiser. They hired a surveyor to plot the squares on the football field. We discovered that the field was narrower on one end by about 7 feet! The cow plopped on a square boundary with an unsold square so the winner only got half. As I recall it was a whole lot of standing around and waiting. Funny for the first 20 minutes, then very, very boring.