Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Mar 29, 2023

• ci : re-enable AVX512 testing (Windows-MSVC) (ggerganov#584)

  Show description for 9cbc404
  anzz1

  authored
  9cbc404

• ggml : init time on first ggml_init() call

  ggerganov

  committed
  b51c717

• llama : fix compile warnings when reading the vocab

  ggerganov

  committed
  0ba76c1

• ggml : add ARM_NEON dequantize_row_q4_1()

  ggerganov

  committed
  cea1c85

• ggml : add ARM_NEON quantize_row_q4_1()

  ggerganov

  committed
  f202ada

• ggml : add ARM_NEON ggml_vec_dot_q4_1()

  ggerganov

  committed
  3b44d30

• rename convert_ggml_to_pth.py -> convert-ggml-to-pth.py (ggerganov#600)

  Show description for 61cbfff
  prusnak

  authored
  61cbfff

• Create chat-13B.bat (ggerganov#592)

  Show description for d9ad104
  

  Royalphax

  and

  anzz1

  authored
  d9ad104

• readme : fix typos

  ggerganov

  authored
  b467702

• readme : add GPT4All instructions (close ggerganov#588)

  ggerganov

  authored
  516d88e

• py : add GPT4All conversion script

  Show description for 53635c0
  ggerganov

  committed
  53635c0

• llama : use the same threshold for OpenBLAS and ggml thread limiting (ggerganov#577)

  Piezoid

  authored
  41318d7

• add example of re-act pattern (ggerganov#583)

  Show description for a6956b2
  tobi

  authored
  a6956b2

• Fix GCC warning about binary literal (ggerganov#595)

  Show description for 83df563
  anzz1

  authored
  83df563

• Fix typo in llama.h (ggerganov#593)

  anzz1

  authored
  a5c42c4

Commits on Mar 28, 2023

• Enable Fused-Multiply-Add (FMA) and F16C/CVT16 vector extensions on MSVC (ggerganov#375)

  Show description for 5a5f8b1
  anzz1

  authored
  5a5f8b1

• CI: fix subdirectory path globbing (ggerganov#546)

  Show description for f121705
  anzz1

  authored
  f121705

• llama : fix linkage with mingw (ggerganov#551)

  Show description for 7f4c5c6
  anzz1

  authored
  7f4c5c6

• ggml : add AVX2 implementation of quantize_row_q4_1 (ggerganov#515)

  Show description for 2a98bc1
  

  slaren

  and

  ggerganov

  authored
  2a98bc1

• py : add temporary script to convert old ggml files to newer version (ggerganov#539)

  Show description for d0aaff5
  

  thement

  and

  jxhor

  authored
  d0aaff5

• py : add capabiliy to convert from ggml back to torch or hf format for further consumption/training/finetuning (ggerganov#403)

  ductai199x

  authored
  d0330fd

• ggml : refactor quantized processing functions (ggerganov#509)

  Show description for 99c5b27
  

  sw

  and

  ggerganov

  authored
  99c5b27

• py : removed unused model variable and verified that the code functions correctly with vocab_only setting. Also confirmed that the code works as expected after running with reduced memory usage…

  Show description for 692ce31
  fritzprix

  authored
  692ce31

• ci : make ctest verbose, hopefully we see what is wrong with the sanitizer

  ggerganov

  committed
  96f9c05

• tests : free llama context at the end of the test

  ggerganov

  committed
  d502bc7

• all : be more strict about converting float to double (ggerganov#458)

  Show description for 436e561
  

  sw

  and

  ggerganov

  authored
  436e561

• deploy : add a Package.swift for SwiftPM support (ggerganov#393)

  Show description for 20e1e84
  j-f1

  authored
  20e1e84

• ggml : introduce structs for the q4 data blocks (ggerganov#356)

  Show description for c1f8850
  

  sw

  and

  ggerganov

  authored
  c1f8850

• gitignore : add "embedding"

  ggerganov

  committed
  e067026

• Check the existence of f16_model_path_base in quantize.py (ggerganov#574)

  Show description for 28ba975
  

  dotpy314

  and

  ryanmiao

  authored
  28ba975

• Fix usage of F16C intrinsics in AVX code (ggerganov#563)

  Show description for a6bdc47
  slaren

  authored
  a6bdc47

• main.cpp fixes, refactoring (ggerganov#571)

  Show description for 7b8dbcb
  anzz1

  authored
  7b8dbcb

• Add embedding example to Makefile (ggerganov#540)

  rjadr

  authored
  4b8efff

Commits on Mar 27, 2023

• Fix missing ggml link in cmake for examples/* on w64-mingw32 (ggerganov#542)

  marcom

  authored
  7e53955

Commits on Mar 26, 2023

• ci: add debug build to sanitizer build matrix (ggerganov#527)

  Green-Sky

  authored
  34c1072

Pagination

PreviousNext