44% on ARC-AGI-1 in 67 cents
A researcher has developed a small, cost-effective transformer model that achieves a 44% score on the ARC-AGI benchmark. The project focuses on improving sample efficiency and reducing computational costs for AI training.
Why it matters
This work challenges the assumption that massive compute resources are required for high-level AI performance, potentially democratizing AI research.
I trained a small transformer from scratch in 1.5hrs on a 5090 Beats many LLMs, and scores the same as TRM/HRM
This is an upgrade to my previous model Faster, better, cheaper and still open source.
This is the 3rd blog in a series of works on ARC-AGI. Prev: Blog 2 , Blog 1 .
Many ppl thought the prev result was impossible. It got attention from top researchers and went viral on X. Eg: Discussions by Lucas Beyer , Jeremy Howard , Rohan Anil , and comments by many others.
I think sample efficiency is the most important problem in AI today and I want to solve it.
The intention behind this work is to (1) find the limits of sample efficiency when restricted to transformers / today’s deep learning methods and (2) reduce costs so iteration is much faster and cheaper.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in