ochafik/llama.cpp

Pull Requests Commits

url: tweak sig of fs_get_cache_file

Olivier Chafik committed 1 year ago

76612a27

url: fs_get_cache_file_path util

Olivier Chafik committed 1 year ago

bc5acfd1

Merge remote-tracking branch 'origin/master' into cache-url

Olivier Chafik committed 1 year ago

92ec18fe

server : smart slot selection using Longest Common Prefix (#7728)

sasha0552 committed 1 year ago

Verified 7a16ce7d

url: save -mu download to new cache location

Olivier Chafik committed 1 year ago

59f7e0d4

vulkan : reuse parent extra for views (#7806)

slaren committed 1 year ago

Verified da799b41

gguf-split : change binary multi-byte units to decimal (#7803)

christianazinn committed 1 year ago

Verified c00fad71

cmake : fix BUILD_SHARED_LIBS=ON build (#7784)

intelmatt committed 1 year ago

Verified 27615f5a

server: update cache_prompt documentation [no ci] (#7745)

JohannesGaessler committed 1 year ago

Verified 7027b27d

server : do not get prompt in infill mode (#7286)

woodx9 committed 1 year ago

Verified a5cabd76

[SYCL] fix softmax r2r result wrong issue (#7811)

pengxin99 committed 1 year ago

Verified d5c938cd

check for nans in imatrix and quantize (#7807)

slaren committed 1 year ago

Verified c9ee7118

server : fix --threads-http arg (#7801)

ggerganov committed 1 year ago

Verified ee459f40

imatrix : migrate to gpt_params (#7771)

ggerganov committed 1 year ago

Verified f83351f9

Added support for . (any character) token in grammar engine. (#6467)

HanClinto committed 1 year ago

Verified ad675e1c

README minor fixes (#7798) [no ci]

mattchediak committed 1 year ago

Verified a143c043

grammars: x{min,max} repetition operator (#6640)

ochafik committed 1 year ago

Verified 55b2d084

llama : add jina v2 base code (#7596)

JoanFM committed 1 year ago

Verified f5d7b268

docker : build only main and server in their images (#7782)

slaren committed 1 year ago

Verified 2d08b7fb

docker : add openmp lib (#7780)

slaren committed 1 year ago

Verified d67caea0

Fix encoding in python scripts (#7733)

Galunid committed 1 year ago

Verified 7672adee

CUDA: refactor mmq, dmmv, mmvq (#7716)

JohannesGaessler committed 1 year ago

Verified 7d1a378b

ggml : refactor rope norm/neox (#7634)

ggerganov committed 1 year ago

Verified 2b338967

readme : remove -ins (#7759)

arch-btw committed 1 year ago

Verified 9973e81c

Fix per token atrributes bits (#7749)

jaime-m-p committed 1 year ago

Verified c90dbe02

Allow number of nodes in CUDA graph to change (#7738)

agray3 committed 1 year ago

Verified b90dc566

common : refactor cli arg parsing (#7675)

ggerganov committed 1 year ago

Verified 1442677f

ggml : remove OpenCL (#7735)

ggerganov committed 1 year ago

Verified 554c247c

llama : remove beam search (#7736)

ggerganov committed 1 year ago

Verified 0cd6bd34

readme : remove obsolete Zig instructions (#7471)

ggerganov committed 1 year ago

Verified 5ca0944a

Older