@santisiri@andycufari m5 macbook pro 128gb ram (aprox 7k), podés correr un 70b deepseek distilled dejando espacio para el OS. Tenés que jugar un poco con el context window, quantification, y si lo corres con llama.cpp or mlx.
it's open source time, with a real leap for world models 🎉
NVIDIA's SANA-WM: a camera-conditioned world model that fits on one GPU. 60s of 720p in 34s on a single 5090 - 2.6B params and Apache 2.0!
https://t.co/Vptws3HDwa
Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.
@atmoio I like everything you say. But AI it's a little more than that right now. It's good at templating ... so if you have strict rules within your company, you can go far.
I've been more and more into the "drafts" and "rewrites". I think doing a draft is writing the spec in code that it's functional, and once you have all the working code, ask the AI to do a rewrite that fits your arquitectural decisions!