Code

Add support for GGUF Phi-3#31844

Code

Merged

SunMarc merged 8 commits intohuggingface:mainhuggingface/transformers:mainfrom a8nova:phi3-ggufa8nova/transformers:phi3-ggufCopy head branch name to clipboardSep 10, 2024

+122-1

Commits

Commits on Sep 8, 2024

• Update docs for GGUF supported models

  a8nova

  committed
  47fb1d1

  Copy full SHA for 47fb1d1

• Add tensor mappings and define class GGUFPhi3Converter

  a8nova

  committed
  58bdb90

  Copy full SHA for 58bdb90

• Fix tokenizer

  a8nova

  committed
  b3d1d93

  Copy full SHA for b3d1d93

• Working version

  a8nova

  committed
  9ed4e2c

  Copy full SHA for 9ed4e2c

• Attempt to fix some CI failures

  a8nova

  committed
  98d70fe

  Copy full SHA for 98d70fe

• Run ruff format

  a8nova

  committed
  21474e1

  Copy full SHA for 21474e1

• Add vocab, merges, decoder methods like LlamaConverter

  a8nova

  committed
  45c7422

  Copy full SHA for 45c7422

• Resolve conflicts since Qwen2Moe was added to gguf

  Show description for 5a7b62e
  a8nova

  committed
  5a7b62e

  Copy full SHA for 5a7b62e