As AI capabilities skyrocket, openness plummets: the scientific community and broader public know little of how frontier models (including open-weight models) are trained. I will describe Marin, a radically new way of doing model development, inspired by true open-source software. Every experiment is done in the open and anyone can suggest ideas, review, and even run experiments through GitHub, providing a better way of doing science that improves on preregistration, reproducibility, and peer review. I will discuss a selection of scientific results that have emerged from Marin, including the best open-source 32B model trained from scratch, rigorous benchmarking of optimizers and data-efficient training recipes. As we scale up, we hope that Marin can be a lab for the open research community to participate in the development of frontier AI.