Skip to content

gpu-mode/popcorn

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 

Repository files navigation

🍿 Popcorn

GPU programming is still too hard, and the current LLMs aren't helping much. We think we can build something better, an LLM that can actually write good GPU code.

We're doing this in public - all our training runs, conversations, and infrastructure will be open source.

Where to Find Us

We mostly talk on discord.gg/gpumode in the popcorn channel. It's pretty casual - drop in when you can.

What We're Working On

Data

We're collecting and synthetically generating as many GPU kernels as we can.

Infrastructure

We need compute - we're building tools to track progress and share results in real-time through Discord.

Research

We're investigating how to prompt, train, deploy, and sample LLMs for effective GPU code generation.

Language Design

We're exploring new abstractions to simplify GPU programming such as ThunderKittens.

Collaborators and Sponsors

Our collaborators include

Our compute sponsors include

And anyone who shares our goals is welcome to join