ggerganov/llama.cpp

Pull Requests Commits
server : allow continue edit on completion mode (#3950)
jhen0409 committed 2 years ago
Verified 4a4fd3ee
Unbreak persimmon after #3837 (#4010)
Galunid committed 2 years ago
Verified df9d1293
scripts: Generalize convert scripts (#3838)
Galunid committed 2 years ago
Verified a75fa576
server : add min_p param (#3877)
Mihaiii committed 2 years ago
Verified 57ad015d
ggml-alloc : fix backend assignments of views (#3982)
slaren committed 2 years ago
Verified 875fb428
gguf : track writer state, free unneeded tensors, cleanup (#3871)
cebtenzzre committed 2 years ago
Verified 0a7c980b
make : do not add linker flags when compiling static llava lib (#3977)
ggerganov committed 2 years ago
Verified 413503d4
ggml : fix backward rope after YaRN (#3974)
xaedes committed 2 years ago
Verified e9c1cecb
Use params when loading models in llava-cli (#3976)
tejom committed 2 years ago
Verified 54b4df88
cuda : supports running on CPU for GGML_USE_CUBLAS=ON build (#3946)
wsxiaoys committed 2 years ago
Verified 46876d2a
llava : expose as a shared library for downstream projects (#3613)
damian0815 committed 2 years ago
Verified 381efbf4
ggml-cuda : fix f16 mul mat (#3961)
slaren committed 2 years ago
Verified 2833a6f6
Allow common process_escapes to handle \x sequences (#3928)
KerfuffleV2 committed 2 years ago
Verified d9ccce2e
server : fix typo for --alias shortcut from -m to -a (#3958)
hoangtamthai committed 2 years ago
Verified bb60fd0b
cuda : fix disabling device with --tensor-split 1,0 (#3951)
cebtenzzre committed 2 years ago
Verified 132d25b8
llama : mark LLM_ARCH_STARCODER as full offload supported (#3945)
wsxiaoys committed 2 years ago
Verified 3d48f42e
cmake : MSVC instruction detection (fixed up #809) (#3923)
netrunnereve committed 2 years ago
Verified c41ea36e
ci : use intel sde when ci cpu doesn't support avx512 (#3949)
netrunnereve committed 2 years ago
Verified a7fac013
cuda : revert CUDA pool stuff (#3944)
slaren committed 2 years ago
Verified 48ade945
gguf-py: Support 01.AI Yi models (#3943)
KerfuffleV2 committed 2 years ago
Verified f28af0d8
metal : round up to 16 to fix MTLDebugComputeCommandEncoder assertion (#3938)
psugihara committed 2 years ago
Verified d9b33fe9
ggml-metal: fix yarn rope (#3937)
jxy committed 2 years ago
Verified 5ba37461
ggml-cuda : move row numbers to x grid dim in mmv kernels (#3921)
slaren committed 2 years ago
Verified abb77e73
speculative : change default p_accept to 0.5 + CLI args (#3919)
ggerganov committed 2 years ago
Verified 8f961abd
common : YAYF (yet another YARN fix) (#3925)
ggerganov committed 2 years ago
Verified 05816027
llama : change yarn_ext_factor placeholder to -1 (#3922)
cebtenzzre committed 2 years ago
Verified 3fdbe6b6
cuda : add ROCM aliases for CUDA pool stuff (#3918)
KerfuffleV2 committed 2 years ago
Verified 629f917c
cmake : fix relative path to git submodule index (#3915)
abetlen committed 2 years ago
Verified 51b2fc11
readme : add notice about #3912
ggerganov committed 2 years ago
Verified 224e7d5b
cuda : fix const ptrs warning causing ROCm build issues (#3913)
ggerganov committed 2 years ago
Verified c7743fe1
Terms Of Use Privacy Policy Refund Policy Impressum