1
Training a 3.8B LLM to 0.384 CORE for $998
A 3.8B model pretrained from scratch to 0.384 CORE on 65B tokens in 43 hours, for $998.
Machine learning engineer and researcher. I train and evaluate deep-learning systems, from language models pretrained from scratch to video understanding.
Visit hugovergnes.github.io → On Hacker News Sep 2026 – Sep 2026
A 3.8B model pretrained from scratch to 0.384 CORE on 65B tokens in 43 hours, for $998.