HipScript
Dev tools
gpu programming playground
SHOW HN2025
How HN took it
10 positive, 7 neutral or asking, 2 mixed, 6 critical, out of 25 comments about the project. Thread score +0.23.
Execution: 5 for, 3 againstIdea: 3 for, 1 againstPerformance: 1 for, 2 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Impressive, clean demo that runs well, especially on Apple devices.
From 4 comments
What an incredible demo/hack. This is actually the simplest way to actually execute CUDA code that I've seen.
bagels on HNNot to mention that it Just Works on Apple devices! Really, really cool.
btown on HNThis is just amazing. So clean and straightforward. On a Mac just run it in Chrome see things work automagically. For real fun change the generations constant to 2000 and delete…
grconner on HNThis is really awesome. Tested it on both Nvidia and Mac GPUs. Interested to know how debugging in a real application would work since WASM is pretty hard to debug and GPU code…
mlepath on HN
Runs on varied GPUs including Mac, Nvidia and AMD RDNA2.
From 2 comments
This is really awesome. Tested it on both Nvidia and Mac GPUs. Interested to know how debugging in a real application would work since WASM is pretty hard to debug and GPU code…
mlepath on HNImpressed that this runs on my RX 6900XT (an RDNA2 GPU) in Chrome without any trouble. Very cool demo, excited to see how people leverage this capability.
uncheckederror on HN
Objections
Blocked on Firefox release and some browsers; feature detection questioned.
From 2 comments
Firefox supports WebGPU, but needs a setting in about:config. I enabled the setting but HipScript still denies running on Firefox with the message: "Please try a Chromium-based…
_nalply on HNOK, I installed Firefox Nightly and enabled dom.webgpu.enabled and gfx.webrender.all corresponding to this article:…
_nalply on HN
Performance questions; unclear it matches native CUDA.
From 3 comments
The GoL example it loaded with seemed to be running way slower than I expected it to. It turns out that there's actually a `usleep(1000 * 100)` call in the code which was…
Ameo on HNHow performant is the CUDA code in browser compared to standalone program? Could we have PyTorch / ML training with CUDA through the browser performing ok?
punnerud on HNWebGPU don't have access to most accelerators in GPU so it would be super slow compared to CUDNN version but pure CUDA would be only x times times faster.
machinekob on HN
Requests
Standalone app that skips WebGPU and runs Vulkan directly.
From 1 comment
I love the idea of this, but sadly I can't get it to work in Firefox, Chrome or Edge on my work pc, probably because I can't find "--enable-features=Vulkan" equivalent in the…
Cieric on HN
Support for running without a detected GPU.
From 1 comment
I don't currently expose the option to run the compiler if a GPU isn't detected, but on systems that do, it exposes a download option that lets you download the SPIR-V kernel to…
lights0123 on HN
Questions
How is this different from web-llm?
From 1 comment
How is this different than web-llm?
bloomingkales on HN
Could PyTorch or ML training run acceptably in the browser?
From 1 comment
How performant is the CUDA code in browser compared to standalone program? Could we have PyTorch / ML training with CUDA through the browser performing ok?
punnerud on HN
How hard is debugging CUDA code running through WASM?
From 1 comment
This is really awesome. Tested it on both Nvidia and Mac GPUs. Interested to know how debugging in a real application would work since WASM is pretty hard to debug and GPU code…
mlepath on HN
Alternatives named
web-llm
From 1 comment
How is this different than web-llm?
bloomingkales on HN