mlc-ai/llm-perf-bench

Issues

• [BUG] mlc-llm benchmark failed with ROCm

  #36 opened a year ago by alexhegit
  1

• Llm perf bench for mobile devices

  #37 opened 10 months ago by BlindDeveloper
  0

• Perplexity and memory use comparisons would be useful

  #9 opened a year ago by JohannesGaessler
  4

• [bug] - crashes when doing build following standard ROCM instructions, related to batching code

  #32 opened a year ago by Sing-Li
  12

• ERROR: Could not build wheels for flash-attn, which is required to install pyproject.toml-based projects

  #31 opened a year ago by VincentXWD
  3

• can 8gb rtx 3060 run the 13b model?

  #13 opened a year ago by hiqsociety
  1

• which repo to download from hugging face for 7b / 13b to try out?

  #12 opened a year ago by hiqsociety
  1

• llama.cpp thread parameter is suboptimal

  #8 opened a year ago by JohannesGaessler
  4

• Docker container seems to be missing Python dependency

  #11 opened a year ago by JohannesGaessler
  4

• llama.cpp compilation settings are suboptimal

  #10 opened a year ago by JohannesGaessler
  7

• tok/sec metric is not clearly defined

  #7 opened a year ago by JohannesGaessler
  1

• CUDA error: no kernel image is available for execution on the device

  #3 opened a year ago by sleepwalker2017
  18

• raise ValueError("Cannot detect local CUDA GPU target!")

  #6 opened a year ago by AegeanYan
  6