Code

Benchmark Code#27

Code

Open

szeyu wants to merge 23 commits intomainEmbeddedLLM/embeddedllm:mainfrom szeyu-benchmark-2EmbeddedLLM/embeddedllm:szeyu-benchmark-2Copy head branch name to clipboard

+856-103

Commits

Commits on Aug 6, 2024

• update new model list with new reuploaded model and ipex option in modelui

  szeyu

  committed
  50f4c34

  Copy full SHA for 50f4c34

• fix the typo of mistral repo id

  szeyu

  committed
  ec2f421

  Copy full SHA for ec2f421

• edit to the latest version of models available

  szeyu

  committed
  dbdefa0

  Copy full SHA for dbdefa0

• change the context length of 128k to 131072

  szeyu

  committed
  0965d51

  Copy full SHA for 0965d51

Commits on Aug 12, 2024

• Merge branch 'main' into szeyu-autoloader-1

  szeyu

  committed
  b905aa9

  Copy full SHA for b905aa9

Commits on Aug 13, 2024

• onnx auto download model if repo id is provided as model path

  szeyu

  committed
  5dbf495

  Copy full SHA for 5dbf495

• formated with black

  szeyu

  committed
  fb2c63e

  Copy full SHA for fb2c63e

Commits on Aug 14, 2024

• fixed with flake8

  szeyu

  committed
  f8c8f27

  Copy full SHA for f8c8f27

• add openvino description and the device gpu

  szeyu

  committed
  d54b4d8

  Copy full SHA for d54b4d8

• update openvino in modelui list

  szeyu

  committed
  1c3b393

  Copy full SHA for 1c3b393

Commits on Aug 15, 2024

• first commit of benchmark code

  szeyu

  committed
  75eff7c

  Copy full SHA for 75eff7c

• update for the markdown to teach about benchmark code usage

  szeyu

  committed
  608670c

  Copy full SHA for 608670c

• Rename benchmark.md to README.md

  szeyu

  authored
  21d95aa

  Copy full SHA for 21d95aa

• Update README.md

  szeyu

  authored
  a038376

  Copy full SHA for a038376

• fixed the bias for encode and output_token_length for openvino

  szeyu

  committed
  ca93ba9

  Copy full SHA for ca93ba9

• Merge branch 'szeyu-benchmark-2' of https://github.com/EmbeddedLLM/embeddedllm into szeyu-benchmark-2

  szeyu

  committed
  79af320

  Copy full SHA for 79af320

• Update loop_ellm_benchmark.py

  szeyu

  authored
  632d651

  Copy full SHA for 632d651

• add prompt bias to fix the token encode margin error for directml

  szeyu

  committed
  e51527d

  Copy full SHA for e51527d

Commits on Aug 16, 2024

• Update ellm_benchmark.py

  szeyu

  authored
  013adc4

  Copy full SHA for 013adc4

• Update loop_ellm_benchmark.py

  szeyu

  authored
  62e0b2c

  Copy full SHA for 62e0b2c

• Update README.md

  szeyu

  authored
  03cc6b7

  Copy full SHA for 03cc6b7

Commits on Sep 2, 2024

• update the benchmark loop to loop without having the model load again in every loop

  szeyu

  committed
  4998e2c

  Copy full SHA for 4998e2c

• Update loop_ellm_benchmark.py

  szeyu

  authored
  769e558

  Copy full SHA for 769e558