Bash one-liners for LLMs(justine.lol)

265 points by severine 2 years ago | 54 comments

Llamafile is the new best way to run a LLM on your own computer(simonwillison.net)

108 points by djha-skin 2 years ago | 22 comments

Using mmap to make LLaMA load faster(justine.lol)

212 points by yawboakye 3 years ago | 133 comments

Faster sorting with SIMD CUDA intrinsics (2024)(winwang.blog)

73 points by winwang 1 year ago | 9 comments

Llamafile for Meltemi: The First LLM for Greek(tselai.com)

42 points by fforflo 2 years ago | 17 comments

Matrix Core Programming on AMD GPUs(salykova.github.io)

96 points by skidrow 1 year ago | 4 comments

Beating NumPy matrix multiplication in 150 lines of C(salykova.github.io)

372 points by p1esk 2 years ago | 69 comments

Blog tutorial implementing high-performance CPU matrix multiplication in about 150 lines of C, following the BLIS design and outperforming NumPy/OpenBLAS on a Ryzen 7700.

The Lost Atomic Update on Loongson CPU(jia.je)

120 points by jiegec 1 month ago | 6 comments

A blog post investigating a hardware erratum on the Loongson LA664 CPU where an atomic instruction loses updates, causing an infinite loop while packaging normaliz for Debian.

Llamafile lets you distribute and run LLMs with a single file(github.com)

972 points by tfinch 2 years ago | 253 comments

GitHub repository for Mozilla-Ocho's llamafile, a tool to distribute and run large language models as a single executable file.

Linux gaming is faster because Windows APIs are becoming Linux kernel features(xda-developers.com)

930 points by haunter 5 months ago | 571 comments

XDA Developers article explaining how Linux gaming performance is improving as Windows API behaviors are implemented as Linux kernel features.

GPUs Go Brrr(hazyresearch.stanford.edu)

1056 points by nmstoker 2 years ago | 243 comments

Stanford Hazy Research blog post on what GPU hardware wants for efficient AI kernels, introducing the ThunderKittens CUDA kernel library.

The Pentium contains a complicated circuit to multiply by three(righto.com)

354 points by Tomte 1 year ago | 138 comments

Ken Shirriff's blog post reverse-engineering the Pentium's ×3 multiplier circuit, used in its base-8 floating-point multiplication, and explaining how it works.