16 comments

  • yu3zhou4 1 day ago
    README is in my opinion (author here) the most interesting - I wrote it to help others build useful mental model to be able to recreate the project yourself, without need to even read my code
    • lukemerrick 1 day ago
      I am not super familiar with C and CUDA, so I read solely for the README and enjoyed it supremely. The blend of cheerful walking through instructive examples and your philosophical takes on how to approach the exercise to get the most out of it put me in a great mood. You captured that special upbeat attitude that comes about when you're doing something as well as you can just because it's so legitimately interesting to you.
    • janalsncm 1 day ago
      Really practical teaching approach. I clicked in to see how safetensors are loaded and just kept reading. Thanks for sharing.
  • dwa3592 1 day ago
    Very nice job on read me.

    >>Physically, LLM is a file which contains a lot of float numbers.

    aka atoms of the LLM.

  • xuanlin314 1 day ago
    The lesson-style README is a great approach. Breaking down LLM inference into digestible steps makes the codebase approachable even for people who haven't touched CUDA before.
  • GoldenJade 1 day ago
    Thanks for sharing this. As someone currently researching LLMs, I'm sure I'll be referencing this quite a bit going forward.
  • tom-wal 1 day ago
    I feel like I learned twice as much in 10 minutes reading this than I did reading LLM for Dummies. Thank you
  • nazgulsenpai 1 day ago
    I love the documentation formatted in lessons. I can't wait to read through it.
  • juancn 1 day ago
    Looks interesting, it reminds me of the first llama.cpp, but better documented.
  • cookiengineer 1 day ago
    Wanted to add that the author has an amazing blog with lots of interesting papers: https://jedrzej.maczan.pl/
  • sylware 1 day ago
    I am looking at a plain and simple C implemented LLM inference, and/or x86_64 assembly implemented, and/or AMD GPU RDNA assembly.

    Anybody?

    • irishcoffee 1 day ago
      I heard once that c++ can become assembly at some point if you type the right things in. :)
      • sylware 6 hours ago
        Well, the whole purpose is to be independent of invisible backdoor injectors...^W I mean compiler, to be more accurate those compilers which deals with computer languages with an absurd and grotesque syntax complexity.
  • einpoklum 1 day ago
    It seems the author believes checking the return values of CUDA API calls is not "tiny" enough :-(
  • smy_smy 1 day ago
    interesting!
  • pslab 1 day ago
    [flagged]
  • alexpandey 1 day ago
    [dead]
  • michaelmjh 11 hours ago
    [dead]
  • aamir_ukmer 14 hours ago
    [dead]
  • harshuljain13 1 day ago
    [dead]