nihui/vkpeak

Issues

• Test fp16 without 16-bit storage

  #3 opened 2 years ago by anholt
  0

• Add INT8-matrix performance - The current industry standard for measuring AI inference in TOPS

  #14 opened a month ago by NikosDi
  0

• Intel's XMX Engine (Tensor cores - FP16 Matrix) support

  #11 opened 6 months ago by NikosDi
  8

• Posting my 4070 results!

  #7 opened 10 months ago by oscarbg
  2

• Docker image

  #12 opened 3 months ago by robertvazan
  0

• ValidationLayer errors on some mesa drivers

  #10 opened 6 months ago by infapi00
  0

• kernel launch latency test..

  #4 opened 2 years ago by oscarbg
  1

• 这实现不是 peak

  #9 opened 10 months ago by tpoisonooo
  4

• vkpeak crashes at vkMapMemory() on APUs with large dedicated GPU memory allocations set in BIOS (>512MB)

  #8 opened 10 months ago by hjc4869
  3

• Test results on RTX 3060 are lower than expected

  #6 opened 2 years ago by edisonchan
  1

• Req: Do not run tests on unsupported hardware

  #5 opened a year ago by Quackdoc
  2

• Crashes on M1 Max

  #2 opened 3 years ago by lbibass
  2

• Int8 & Int64 support?

  #1 opened 3 years ago by oscarbg
  1