Hacker GitHub

← All repos

danieldk/gemm-benchmark

View on GitHub ↗ ★ 11 Rust

Simple [sd]gemm benchmark, similar to ACES dgemm

10 mentions on Hacker News by 1 person: 0 stories, 10 comments · first 2021-07-22 · latest 2023-07-20

Mentions per year 10 mentions · busiest year 2023 (5)
20082026
AllStoriesComments NewestTopOldest
  1. …The matrix multiplication on the M1 is on par with ~10 threads/cores of Ryzen 5900X. [1] https://github.com/danieldk/gemm-benchmark#example-results
    comment by danieldk · 2023-07-20 · in Running Stable Diffusion in 260MB of RAM
  2. …incur the cost of data transfers from main memory to GPU memory and vice versa. [1] Benchmarks: https://github.com/danieldk/gemm-benchmark#example-results https://explosion.ai/blog/metal-performance-shaders (scroll down a bit for AMX and…
    comment by danieldk · 2023-05-27 · in Ask HN: What is an A.I. chip and how does it work?
  3. …or faster at a fraction of the wattage. See eg. https://explosion.ai/blog/metal-performance-shaders https://github.com/danieldk/gemm-benchmark#1-to-16-threads
    comment by danieldk · 2023-05-16 · in Apple previews Live Speech, Personal Voice, and more new accessibility features
  4. …performance is roughly twice that of the M1 Pro/Max and four times that of the M1. Benchmarks: https://github.com/danieldk/gemm-benchmark#1-to-16-threads
    comment by danieldk · 2023-01-05 · in How to get 1.5 TFlops of FP32 performance on a single M1 CPU core
  5. …faster. See here for sgemm benchmarks for everything from the M1 to M1 Ultra and 1 to 16 threads: https://github.com/danieldk/gemm-benchmark#1-to-16-threads
    comment by danieldk · 2023-01-05 · in How to get 1.5 TFlops of FP32 performance on a single M1 CPU core
  6. …with a high-end Ryzen CPU last year, and builds where much slower than even the vanilla M1.) [1] https://github.com/danieldk/gemm-benchmark#1-to-16-threads
    comment by danieldk · 2022-03-24 · in AWS Graviton vs. M1 vs. M1 Pro Node.js Benchmarks
  7. …on non-specialized hardware. E.g., a Ryzen 5900X, can do ~150 GFLOP/s single-threaded with MKL. [1] https://github.com/danieldk/gemm-benchmark#1-to-16-threads
    comment by danieldk · 2022-01-25 · in WebAssembly techniques to speed up matrix multiplication
  8. …calling Accelerate in some way) is just a huge waste of everyone's CPU cycles and electricity. [1] https://github.com/danieldk/gemm-benchmark#1-to-16-threads
    comment by danieldk · 2022-01-25 · in WebAssembly techniques to speed up matrix multiplication
  9. …blow this away. Here are some of my single precision GEMM benchmarks for the M1 and M1 Pro: https://github.com/danieldk/gemm-benchmark#1-to-16-threads tl;dr, the M1 can do ~1300 GFLOP/s and the M1 Pro up to ~2700GFLOP/s. On the vanilla M1,…
    comment by danieldk · 2022-01-25 · in WebAssembly techniques to speed up matrix multiplication
  10. …1.02% gemm-benchmark libmkl_avx2.so.1 [.] mkl_blas_avx2_sgemm_scopy_right4_ea Benchmarked using https://github.com/danieldk/gemm-benchmark and oneMKL 2021.3.0.
    comment by danieldk · 2021-07-22 · in Intel Distribution for Python