ggerganov/llama.cpp

Pull Requests Commits

Fix typo in llama.h (#593)

anzz1 committed 2 years ago

Verified a5c42c4b

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

anzz1 committed 2 years ago

Verified 5a5f8b15

CI: fix subdirectory path globbing (#546)

anzz1 committed 2 years ago

Verified f1217055

llama : fix linkage with mingw (#551)

anzz1 committed 2 years ago

Verified 7f4c5c66

ggml : add AVX2 implementation of quantize_row_q4_1 (#515)

slaren committed 2 years ago

Verified 2a98bc18

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

thement committed 2 years ago

Verified d0aaff57

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

ductai199x committed 2 years ago

Verified d0330fd7

ggml : refactor quantized processing functions (#509)

sw committed 2 years ago

Verified 99c5b276

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 due to deletion of no-longer-needed variable. (#547)

fritzprix committed 2 years ago

Verified 692ce316

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

ggerganov committed 2 years ago

Verified 96f9c050

tests : free llama context at the end of the test

ggerganov committed 2 years ago

Verified d502bc7c

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

sw committed 2 years ago

Verified 436e5619

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

j-f1 committed 2 years ago

Verified 20e1e848

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

sw committed 2 years ago

Verified c1f88506

gitignore : add "embedding"

ggerganov committed 2 years ago

Verified e0670260

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

dotpy314 committed 2 years ago

Verified 28ba975a

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

slaren committed 2 years ago

Verified a6bdc47c

main.cpp fixes, refactoring (#571)

anzz1 committed 2 years ago

Verified 7b8dbcb7

Add embedding example to Makefile (#540)

rjadr committed 2 years ago

Verified 4b8efff0

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

marcom committed 2 years ago

Verified 7e539557

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

Green-Sky committed 2 years ago

Verified 34c1072e

Fix undefined variables in debug build, remove unused variables (#531)

sw committed 2 years ago

Verified 939ad2d3

Add support for linux/arm64 platform during Docker Builds (#514)

gaby committed 2 years ago

Verified 8c2ec5e2

Update README and comments for standalone perplexity tool (#525)

sw committed 2 years ago

Verified b391579d

[main] fix infinite generation (-n == -1) (#523)

anzz1 committed 2 years ago

Verified 7a87d31f

Add logo to README.md

ggerganov committed 2 years ago

Verified 348d6926

Exit from interactive mode if input stream is bad (#491)

haraldF committed 2 years ago

Verified 33e35b8f

CI: Run other sanitizer builds even if one fails (#511)

anzz1 committed 2 years ago

Verified 19726169

Clarify console output in convert-pth-to-ggml.py (#512)

jp-x-g committed 2 years ago

Verified f732695c

CMake / CI additions (#497)

anzz1 committed 2 years ago

Verified 2f7bf7dd

Newer Older