How to Beat Unsloth's CUDA Kernel Using Mojo–With Zero GPU Experience
1–2 of 2 posts
Re: How to Beat Unsloth's CUDA Kernel Using Mojo–With Zero GPU Experience
#2David Robertson took a quantization challenge designed for CUDA experts, and solved it in Mojo with AI assistance, and ended up 1.07x to 1.84x faster than the state-of-the-art C++/CUDA implementation.