I'm suuuper late to the party, I know, but I was pleasantly surprised by how good the 3B Llama3.2 model is to use on my 16GB 2020 macbook M1, via Ollama and @Simon Willison's stellar โ€œllmโ€ package. I'm referring to this here:
Llama 3.2 goes small and multimodal ยท Ollama Blog
Ollama partners with Meta to bring Llama 3.2 to Ollama.
 And this llm package:
LLM: A CLI utility and Python library for interacting with Large Language Models
 It feels comparable to the speed of using the lightweight models on Perplexity.ai, and it feels fast enough to me just run queries locally. Really nice piece of CLI design - it's a real pleasure to use.
Dear fediverse nerds: If you're interested in a more open, diverse market like me, then you might find the 6th International Open Search Symposium 2024, (also known as #ossym24 to its friends) interesting. I'm going for next week, and I'll be in Munich from Wednesday to Friday. I've written a bit more about why I'm going:
Green Web Foundation
Meet us at the 6th International Open Search Symposium in Munich in October - Green Web Foundation
Our executive director Chris will be in Munich this October, for the 6th International Open Search Symposium. Read on to find out what it is, and w...
 /1