Embed this proof
Contributed CUDA kernel optimization reducing inference latency by 34%
<iframe src="https://powstik.com/embed/proof/5193b5eb-dbb4-4a22-87c3-5a8579f6e70a" width="100%" height="220" frameborder="0" style="border-radius:12px;overflow:hidden"></iframe>
Add this to your personal site, Notion, or portfolio →
← Back to proof