Activeloop (YC S18) is hiring a VP of Engineering in Mountain View (on-site) (careers.activeloop.ai)

Back in the old days, people used to do general-purpose GPU programming by using shaders like GLSL. This is what inspired NVIDIA (and other companies) to eventually create CUDA (and friends). This is an implementation of GPT-2 using WebGL and shaders. Enjoy!

Comments (25)

ianand · 1d ago

As the guy who did GPT2 in Excel, very cool and kudos!!

Curious why you chose WebGL over WebGPU? Just to show it can be done?

(Also see my other comment about fetching weights from huggingface)

nathan-barry · 1d ago

This was a final project for a graphics class where we used WebGL a lot. Also I was just more familiar with OpenGL and haven't looked that much into webGPU

divan · 22h ago

Someone needs to implement Excel using graphics shaders now.

ianand · 11h ago

Lol. You joke but... https://www.reddit.com/r/GraphicsProgramming/comments/pg9dw8...

ronsor · 1d ago

Probably because WebGPU support is still rather iffy.

_dijs · 1d ago

ianand, I immediately thought of you when I saw this post. Miss you friend.

ianand · 11h ago

Dude, been forever. Thanks. Will DM you.

littlestymaar · 1d ago

> Curious why you chose WebGL over WebGPU? Just to show it can be done?

For a WebGPU implementation, one can use transformers.js directly (or many other libraries actually), maybe WebGL is more original.

[1]: https://huggingface.co/docs/transformers.js/index

85392_school · 23h ago

Transformers.js wraps the ONNX runtime which is rather versatile (WASM, WebGL, WebGPU, and WebNN). It's not the backend that makes it novel.

Philpax · 1d ago

I'm reminded of this VRChat world, which runs Qwen2-0.5B in a shader: https://vrchat.com/home/launch?worldId=wrld_e919c1f4-8399-43...

supuun · 5h ago

> Back in the old days, people used to do general-purpose GPU programming by using shaders like GLSL. This is what inspired NVIDIA (and other companies) to eventually create CUDA (and friends).

I learned about this at the lecture last week. So fun to see this project.

gitroom · 1d ago

Nice, that takes some serious grinding to pull off in shaders. Makes me kinda wonder - does pushing old tools like WebGL actually force deeper understanding of how these models work compared to just grabbing a new library?

helloplanets · 1d ago

Would absolutely read a 'making of' post on this project!

badmonster · 1d ago

Super cool project! How much of the inference time is actually GPU-bound vs CPU-bound in the browser? Curious if WebGPU would offer a big perf boost over WebGL2 here.

SamosaGuru · 1d ago

Would be really cool to have it hosted on a website if it can be a static page so we could try it out

nathan-barry · 1d ago

Working on getting it up on Github pages, will let you know. Not sure if there will be an issue with hosting the weights, I'll look into it

Philpax · 1d ago

You may be able to fetch the weights directly from Hugging Face. I'd try that first.

ianand · 1d ago

Checkout https://github.com/jseeio/gpt2-tfjs fetches the weights for GPT2 from huggingface on the fly.

jasonjmcghee · 1d ago

I haven't looked at your implementation, but if you can read ONNX weights... https://huggingface.co/Xenova/gpt2

jasonjmcghee · 1d ago

Incredible. Was going to suggest tossing up a github page with it, but looks like one is currently in-progress? (https://nathan-barry.github.io/gpt2-webgl/) but doesn't seem to load for me currently.

Looks like it's trying to load the typescript directly / hosting the source files: `<script type="module" src="src/main.ts"></script>` - instead of hosting the dist folder.

nathan-barry · 1d ago

Yeah, right now the weights aren't even pushed to the repo so there's no way for the github page to load them in. I've looked a tiny bit into it and might end up making a fix for it, having it load the weights by fetching them from somewhere else on load, but probably not today.

There is a link to the zipped up folder with the weights here: https://github.com/nathan-barry/gpt2-webgl/releases/download...

If anyone wants to make a PR which fetches the file from there, unzips it, then loads in the weights from there that would be greatly appreciated.

Narishma · 1d ago

Now do it with the fixed function pipeline so I can run it on my 3dfx Voodoo card.

throwaway314155 · 1d ago

This is pretty rad. I've built and run the project on my MacBook and can't seem to get the visualizations to show as they appear in the GitHub demonstration. I'm running Firefox but I think that should support WebGL features you're using?

Happy to file an issue on the repository as well if you'd prefer that.

nathan-barry · 1d ago

Yeah it'd be great if you could file an issue and show what you see! I was using Safari but I'm gonna check other browsers to test if it works on them

throwaway314155 · 1d ago

Great! I've filed an issue.

Stellar Sleep (YC S23) is hiring a product engineer in SF (ycombinator.com)

OneText (YC W23) Is Hiring a DevOps/DBA Lead Engineer

Toma (YC W24) Is Hiring Engs #3-4 (AI for Automotive) (ycombinator.com)

Waypoint Transit (YC W25) is hiring a software engineer (workatastartup.com)

GroMo (YC W21) Is Hiring (ycombinator.com)

Archil (YC F24) Is Hiring a Distributed Systems Engineer (In-Person, SF)

Modern Realty (YC S24) Is Hiring (workatastartup.com)

Hestus, Inc. (YC S24) Is Hiring an ML Engineer to Revolutionize CAD (ycombinator.com)

Activeloop (YC S18) is hiring a VP of Engineering in Mountain View (on-site) (careers.activeloop.ai)

Optery (YC W22) – Engineering Team Lead and Engineers with Node.js (U.S., Latam) (jobs.ashbyhq.com)

Extend (YC W23) is hiring engineers to build LLM document processing (jobs.ashbyhq.com)

Parity (YC S24) is hiring founding engineers to build an AI SRE (in-person, SF) (ycombinator.com)

Freshpaint (YC S19) is hiring back end and front end engineers (Remote, US only)

MobileBoost (YC S21) Is Hiring a Founding Back End/Platform Engineer (Remote) (ycombinator.com)

Gym Class (YC W22) Is Hiring Character Animation Engineering Lead (ycombinator.com)

Foundry (YC F24) is hiring – Come build a world model for the web

Bild AI (YC W25) is hiring a founding engineer in SF (ycombinator.com)

Tenjin (YC S14) Is Hiring a Senior Ad Attribution Engineer (Ruby, Go) (ycombinator.com)

Onyx (YC W24) Is Hiring for ML Engineer (ycombinator.com)

Recover (YC W21) Is Hiring (ycombinator.com)

GiveCampus (YC S15) Is Hiring Sr engineers passionate about education (givecampus.breezy.hr)

Cekura (Formerly Vocera) (YC F24) Is Hiring (ycombinator.com)

Spark AI (YC W24) is hiring a full-stack engineer in San Francisco (ycombinator.com)

FurtherAI (YC W24) Is Hiring Software and AI Engineers (ycombinator.com)

Weave (YC W25) is hiring a founding engineer (ycombinator.com)

Infisical (YC W23) Is Hiring Design Engineer in San Francisco (ycombinator.com)

Dot (YC S21) is hiring a sales engineer to automate analytics (fully remote) (ycombinator.com)

Mux (YC W16) is hiring engineering managers for video at scale (mux.com)

EasyPost (YC S13) Is Hiring (easypost.com)

Tesorio (YC S15) Is Hiring a Senior Back End Engineer in Latam (100% Remote) (tesorio.com)

Formance (YC S21) Is Hiring Engineers to Build OSS Financial Infrastructure (ycombinator.com)

Show HN: GPT-2 implemented using graphics shaders

Comments (25)