
Academic lab · Stanford, United States
Hazy Research
A Stanford research group led by Chris Re, known for efficient architectures and GPU kernels such as ThunderKittens, Hyena, Based and Monarch Mixer.
Stanford, United States
- Type
- Academic lab
- Based in
- Stanford, United States
- Key papers
- 4
- Featured projects
- 7
About
Hazy Research is a Stanford computer science group led by Professor Chris Re. It studies learning from weak supervision and the mathematical foundations of these methods, and turns the results into systems that help people build and maintain machine learning models.
Its blog covers GPU kernel programming with ThunderKittens, HipKittens and ParallelKittens, efficient architectures including Monarch Mixer, Based, Hyena, S4 and FlashFFTConv, local on-device models such as Minions and OpenJarvis, and long-context memory techniques such as Cartridges.
Mission and ambitionsThe group aims to make machine learning systems faster and easier to build by pairing theory with open implementations.
Known for
- 01
ThunderKittens GPU kernels
- 02
Hyena, Based and Monarch Mixer architectures
- 03
Minions local-model work
Open source
Tile primitives for speedy kernels.
Knowledge base construction system.
Big and small LLMs working together.
Data-centric AI resources.
- based-1bmodel
Based 1B model.
- based-swdedataset
SWDE for Based evaluation.
- based-fdadataset
FDA for Based evaluation.
Research papers
Notable releases
Openness
Open code and weights.
People
- Chris RéProfessor and group lead
- Kelly BuchananPostdoctoral researcher
- Jon Saad-FalconPhD student
- Divya NoriPhD student
- Roberto GarciaPhD student
Listed from the lab's own pages and public profiles.
Contact and community
Contact details and links are on the lab's website (https://hazyresearch.stanford.edu); code questions go through issues on their GitHub organisation.
Sources
Compiled from the lab's public pages, Hugging Face and GitHub. Last checked 2026-11-17. Something wrong or out of date? Tell us on Discord.



