ochafik/llama.cpp

Pull Requests Commits

Makefile: move unused logits flags where they don't interfere w/ targets

ochafik committed 2 years ago

58bb7d56

Merge remote-tracking branch 'origin/master' into skip-unused-2

ochafik committed 2 years ago

55d49b26

readme : update hot topics

ggerganov committed 2 years ago

Verified 7eb41179

llama : allow gguf RoPE keys to be overridden with defaults (#3240)

cebtenzzre committed 2 years ago

Verified a5661d7e

benchmark-matmult : do not use integer abs() on a float (#3277)

cebtenzzre committed 2 years ago

Verified 65c2c1c5

flake : Restore default package's buildInputs (#3262)

tpdns90321 committed 2 years ago

Verified 80834dae

CI: FreeBSD fix (#3258)

alonfaraj committed 2 years ago

Verified a40f2b65

examples : fix benchmark-matmult (#1554)

ggerganov committed 2 years ago

Verified d119c04c

make : restore build-info.h dependency for several targets (#3205)

cebtenzzre committed 2 years ago

Verified 8781013e

ci : switch cudatoolkit install on windows to networked (#3236)

Green-Sky committed 2 years ago

Verified 7ddf1855

CUDA: fix peer access logic (#3231)

JohannesGaessler committed 2 years ago

Verified ee66942d

CUDA: enable peer access between devices (#2470)

JohannesGaessler committed 2 years ago

Verified 111163e2

llama.cpp : show model size and BPW on load (#3223)

slaren committed 2 years ago

Verified 8b428c9b

CUDA: fix scratch malloced on non-main device (#3220)

JohannesGaessler committed 2 years ago

Verified 578d8c8f

Enable BUILD_SHARED_LIBS=ON on all Windows builds (#3215)

IsaacDynamo committed 2 years ago

Verified b541b4f0

Enable build with CUDA 11.0 (make) (#3132)

Microflame committed 2 years ago

Verified 5dbc2b32

Fixing the last deviations from sentencepiece indicated by test-tokenizer-1 (#3170)

goerch committed 2 years ago

Verified b08e75ba

examples : add compiler version and target to build info (#2998)

cebtenzzre committed 2 years ago

Verified e6616cf0

check C++ code with -Wmissing-declarations (#3184)

cebtenzzre committed 2 years ago

Verified 3aefaab9

fix build numbers by setting fetch-depth=0 (#3197)

cebtenzzre committed 2 years ago

Verified 69eb67e2

llama : add support for StarCoder model architectures (#3187)

wsxiaoys committed 2 years ago

Verified 4fe09dfe

common : do not use GNU zero-length __VA_ARGS__ extension (#3195)

cebtenzzre committed 2 years ago

Verified 80291a1d

metal : fix bug in soft_max kernels (out-of-bounds access) (#3194)

ggerganov committed 2 years ago

Verified c6f1491d

convert : make ftype optional in simple scripts (#3185)

cebtenzzre committed 2 years ago

Verified e3d87a6c

sync : ggml (Metal F32 support + reduce ggml-alloc size) (#3192)

ggerganov committed 2 years ago

Verified 8c00b7a6

cmake : fix building shared libs for clang (rocm) on windows (#3176)

Engininja2 committed 2 years ago

Verified 7e50d34b

flake : use pkg-config instead of pkgconfig (#3188)

kurnevsky committed 2 years ago

Verified 235f7c19

metal : relax conditions on fast matrix multiplication kernel (#3168)

ggerganov committed 2 years ago

Verified a51b6876

cmake : fix llama.h location when built outside of root directory (#3179)

abetlen committed 2 years ago

Verified 76164fe2

ci : Cloud-V for RISC-V builds (#3160)

alitariq4589 committed 2 years ago

Verified c2ab6fe6

Older