ggerganov/llama.cpp

Pull Requests Commits
Server: add tests for batch size, different seeds (#6950)
JohannesGaessler committed 2 years ago
Verified 3ea0d360
convert-hf : flake8 linter doesn't like semicolons
compilade committed 2 years ago
56f60f5d
CUDA: CUDART < 11.7 workaround for __hmax, __hmax2 (#7019)
JohannesGaessler committed 2 years ago
Verified 1613ef8d
ci : exempt confirmed bugs from being tagged as stale (#7014)
slaren committed 2 years ago
Verified c4ec9c0d
convert-hf : simplify MoE weights stacking
compilade committed 2 years ago
cde9ea65
perplexity: more statistics, added documentation (#6936)
JohannesGaessler committed 2 years ago
Verified a8f9b076
convert-hf : remove unused n_dims in extra_*_tensors
compilade committed 2 years ago
698f0b34
convert : upgrade to sentencepiece v0.2.0
compilade committed 2 years ago
c33775bc
Merge branch 'master' into compilade/convert-hf-refactor
compilade committed 2 years ago
0d720acb
convert-hf : begin refactoring write_tensor
compilade committed 2 years ago
47e02eb7
switch to using localizedDescription (#7010)
bakkot committed 2 years ago
Verified f364eb6f
metal : remove deprecated error code (#7008)
ggerganov committed 2 years ago
Verified 77e15bec
metal : log more info on error (#6987)
bakkot committed 2 years ago
Verified a68a1e7e
ggml : add Flash Attention (#5021)
ggerganov committed 2 years ago
Verified 9c67c277
convert : use utf8 encoding (#7000)
ggerganov committed 2 years ago
Verified 952d03db
Improve usability of --model-url & related flags (#6930)
ochafik committed 2 years ago
Verified 8843a98c
Extending grammar integration tests (#6644)
HanClinto committed 2 years ago
Verified b8c1476e
main : fix typo in comment in main.cpp (#6985)
danbev committed 2 years ago
Verified 5539e6fd
build(cmake): simplify instructions (`cmake -B build && cmake --build build ...`) (#6964)
ochafik committed 2 years ago
Verified b8a7a5a9
ci : tmp disable gguf-split (#6983)
ggerganov committed 2 years ago
Verified d2c898f7
ggml : fix __MSC_VER -> _MSC_VER (#6977)
ggerganov committed 2 years ago
Verified 544f1f10
llava-cli : multiple images (#6969)
cpumaxx committed 2 years ago
Verified ffe66657
readme : update hot topics
ggerganov committed 2 years ago
Verified 24affa7d
llama : fix BPE pre-tokenization (#6920)
ggerganov committed 2 years ago
Verified f4ab2a41
sampling : use std::random_device{}() for default random seed (#6962)
dwrensha committed 2 years ago
Verified 3f167476
convert : fix conversion of some BERT embedding models (#6937)
christianazinn committed 2 years ago
Verified 3055a418
make : change GNU make default CXX from g++ to c++ (#6966)
przemoc committed 2 years ago
Verified 577277ff
ci : add building in MSYS2 environments (Windows) (#6967)
przemoc committed 2 years ago
Verified ca7f29f5
llama : fix typo LAMMAFILE -> LLAMAFILE (#6974)
JohannesGaessler committed 2 years ago
Verified c4f708a9
Fix more int overflow during quant (PPL/CUDA). (#6563)
dranger003 committed 2 years ago
Verified e00b4a8f
Terms Of Use Privacy Policy Refund Policy Impressum