Hackernews posts about CUDA

CUDA is a parallel computing platform and programming model developed by NVIDIA that allows developers to leverage the processing power of graphics processing units (GPUs) for general-purpose computation.

1. Writing Speed-of-Light Flash Attention for 5090 in CUDA C++ (gau-nernst.github.io)
   159 points by dsr12 26 days ago | 34 comments
2. Show HN: An ncurses CUDA-based fluid simulation (github.com)
   97 points by goosethe 22 days ago | 12 comments
3. Accelerated Game of Life with CUDA / Triton (www.boristhebrave.com)
   49 points by BorisTheBrave 7 days ago | 6 comments
4. Show HN: Flox – Nvidia CUDA available for the Nix ecosystem (flox.dev)
   14 points by ronef 8 days ago | discuss
5. Nvidia CUDA now available via OS and package managers (developer.nvidia.com)
   12 points by Areibman 8 days ago | discuss
6. Llama.cpp: Deterministic Inference Mode (CUDA): RMSNorm, MatMul, Attention (github.com)
   6 points by diwank 3 days ago | discuss
7. Oakridge National Lab CUDA Training Series (www.olcf.ornl.gov)
   5 points by pbd 3 days ago | discuss
8. Canonical announces it will support and distribute Nvidia CUDA in Ubuntu (canonical.com)
   5 points by giuliomagnifico 3 days ago | discuss
9. Flox, the Nix Foundation, and Nvidia Partner for CUDA (flox.dev)
   5 points by Bogdanp 8 days ago | discuss
10. CUDA 13 release notes [pdf] (docs.nvidia.com)
    5 points by tzury 13 days ago | discuss
11. You can get Nvidia's CUDA on three popular enterprise Linux distros now (www.zdnet.com)
    4 points by CrankyBear 3 days ago | 1 comments
12. Nvidia CUDA Toolkit 13.0 Is Out (www.servethehome.com)
    4 points by PaulHoule 27 days ago | discuss
13. 10x Your CUDA Productivity: Python for Programming the GPU (www.nvidia.com)
    3 points by pjmlp 20 days ago | discuss
14. From Zero to GPU: A Guide to Building and Scaling Production-Ready CUDA Kernels (huggingface.co)
    3 points by wavelander 29 days ago | discuss
15. Parrallel String Matching on CUDA (ieeexplore.ieee.org)
    2 points by cwmoore 22 days ago | 1 comments
16. Show HN: Paragon: A Go-native AI framework with WebGPU/Vulkan (no CUDA lock-in) (openfluke.com)
    2 points by openfluke 23 days ago | 1 comments
17. From Zero to GPU: A Guide to Building and Scaling Production-Ready CUDA Kernels (huggingface.co)
    2 points by danieldk 14 days ago | discuss
18. Why are CUDA kernels hard to optimize? (www.johndcook.com)
    2 points by ibobev 21 days ago | discuss
19. Show HN: I Parallelized RNN Training from O(T) to O(log T) Using CUDA (dhruvmsheth.github.io)
    1 points by omegablues about 3 hours ago | discuss
20. Accelerated Game of Life with CUDA / Triton (www.boristhebrave.com)
    1 points by ibobev 6 days ago | discuss
21. Show HN: Luminal – Open-source, search-based GPU compiler (github.com)
    153 points by jafioti 29 days ago | 60 comments
22. Show HN: RunMat – a V8 inspired Rust runtime for the landlocked Matlab language (runmat.org)
    15 points by nallana 28 days ago | 6 comments
23. Show HN: GitArsenal – One-click setup for any GitHub repository (www.gitarsenal.dev)
    6 points by rs545837 28 days ago | 3 comments
24. Clear Thinking beats deep thinking in LLM models
    3 points by sanskarmishra 3 days ago | 4 comments
25. H100 PCIe – 1.86 TB/s memcpy roofline and 8× uplift
    3 points by GPUrouter 5 days ago | discuss
26. Show HN: Omni-LPR – A multi-protocol license plate recognition server
    1 points by habedi0 17 days ago | discuss
27. Cuba's electrical grid collapses in nationwide blackout (www.reuters.com)
    30 points by petethomas 8 days ago | 9 comments
28. Gram: Open-Source platform to create, curate and host MCP servers (github.com)
    20 points by marclave 9 days ago | 2 comments
29. Cuba entirely without power following electric grid collapse (www.nbcnews.com)
    13 points by bookofjoe 5 days ago | 1 comments
30. The Cudy AX3000 Wi-Fi 6 System (With OpenWRT) (taoofmac.com)
    6 points by rcarmo 4 days ago | discuss