ochafik/llama.cpp

Pull Requests Commits

gguf: add GGUFReader.read_field(field) method + read template example

ochafik committed 1 year ago

0d47c43a

Reset schedule earlier to allow overlap with ggml graph computation on device (#6933)

agray3 committed 1 year ago

Verified 928e0b70

quantize: add imatrix and dataset metadata in GGUF (#6658)

phymbert committed 1 year ago

Verified 0c4d489e

add basic tensor data validation function (#6884)

slaren committed 1 year ago

Verified 017e6999

gguf : fix mismatch between alloc and free functions (#6929)

slaren committed 1 year ago

Verified e2764cd7

llamafile : use 64-bit integers in sgemm (#6928)

jart committed 1 year ago

Verified 4b1c3c98

ci: server: fix python installation (#6925)

phymbert committed 1 year ago

Verified bbe3c6e7

server: stop generation at `n_ctx_train` if `n_predict` is not set (#6638)

phymbert committed 1 year ago

Verified 7f5ff558

ci: server: fix python installation (#6922)

phymbert committed 1 year ago

Verified 9e4e077e

Merge pull request from GHSA-p5mv-gjc5-mwqv

ggerganov committed 1 year ago

Verified 83b72cb0

ci: server: fix python installation (#6918)

phymbert committed 1 year ago

Verified d4a9afc1

ci: fix concurrency for pull_request_target (#6917)

phymbert committed 1 year ago

Verified 7d641c26

bench: server add stop word for PHI-2 (#6916)

phymbert committed 1 year ago

Verified 5790c8da

llava : add support for moondream vision language model (#6899)

vikhyat committed 1 year ago

Verified 46e12c46

cmake : restore LLAMA_LLAMAFILE_DEFAULT

ggerganov committed 1 year ago

Verified dba497e0

cmake : remove obsolete ANDROID check

ggerganov committed 1 year ago

Verified fa0b4ad2

llama : synchronize before get/set session data (#6911)

slaren committed 1 year ago

Verified d6e1d44f

ci : tmp disable slow tests

ggerganov committed 1 year ago

Verified 853d06ff

readme : update model list (#6908)

BarfingLemurs committed 1 year ago

Verified 3fe0596c

llama : check that all the tensor data is in the model file (#6885)

slaren committed 1 year ago

Verified 0ead1f10

ggml : fix redefinition of vaddvq_f32 for 32-bit ARM (#6906)

ggerganov committed 1 year ago

Verified 51543729

clip : rename lerp function to avoid conflict (#6894)

danbev committed 1 year ago

Verified 4ab99d8d

ggml : fix MIN / MAX macros (#6904)

ggerganov committed 1 year ago

Verified 54770413

tests : minor bash stuff (#6902)

ggerganov committed 1 year ago

Verified aa750c1e

quantize : add '--keep-split' to quantize model into shards (#6688)

zj040045 committed 1 year ago

Verified 1966eb26

README: add graphic for matrix multiplication (#6881)

JohannesGaessler committed 1 year ago

Verified 784e11de

llama : add llama_get_pooling_type function (#6862)

iamlemec committed 1 year ago

Verified b4e4b8a9

server : do not apply Markdown formatting in code sections (#6850)

mgroeber9110 committed 1 year ago

Verified 3fe847b5

common : revert showing control tokens by default for server (#6860)

K-Mistele committed 1 year ago

Verified 37246b10

Server: fix seed for multiple slots (#6835)

JohannesGaessler committed 1 year ago

Verified 28103f48

Older