https://prismml.com/news/bonsai-27bThis #LLM took an hour at 0.7 tokens per second to make me a little website, because its running on a regular laptop (not made by Apple).The result approximates qwen-3.7-27b in quality, which is the impressive part.This illustrates both how some #ai tasks will probably never be able to be done locally (because we believe in the laws of physics), but also how far you can push efficiency with clever ML science (im assuming we are hitting diminishing returns somewhere around here).#localai #aiResearch #digitalAutonomy
Related
Gemini Spark can now do more things for you and is faster at doing themIt's also expanding to more regions.https://www.a...
Gemini Spark can now do more things for you and is faster at doing themIt's also expanding to more regions.https://www.androidauthority.com/gemini-spark-workspace-features-speed-im...
🥸 Can a hand-drawn mustache break the internet? AI is powerful, but without engineering standards, it’s surprisingly fra...
🥸 Can a hand-drawn mustache break the internet? AI is powerful, but without engineering standards, it’s surprisingly fragile. From the "PDF maze" to $650B data centers, let's talk ...
Selfware, software that fits you instead of the worldA blog by @erikjPeople need not only to obtain things; they need ab...
Selfware, software that fits you instead of the worldA blog by @erikjPeople need not only to obtain things; they need above all the freedom to make things among which they can live...