Show HN: High speed graphics rendering research with tinygrad/tinyJIT
quantbagel Thursday, January 22, 2026
I saw a tweet that tinygrad is so good that you could make a graphics library that wraps tg. So I’ve been hacking on a gtinygrad, and honestly it convinced me it could be used for legit research.
The JIT + tensor model ends up being a really nice way to express light transport all in simple python, so I reimplemented some new research papers from SIGGRAPH like REstir PG and SZ and it just works. instead of complicated cpp its just a 200 LOC of python.
Summary
The article describes the development of gtinygrad, a minimalist deep learning library designed for educational and research purposes. It provides a simple and easy-to-understand implementation of basic neural network components, allowing users to experiment with and understand the inner workings of deep learning models.
20
8
Summary
github.com