ggerganov/llama.cpp

Pull Requests Commits

llama.swiftui : improve bench

ggerganov committed 1 year ago

Verified 86506662

llama : remove "mostly" from model infos

ggerganov committed 1 year ago

Verified 5c5bdba6

llama.swiftui : remove model from project

ggerganov committed 1 year ago

Verified 262fd466

llama.swiftui : avoid data copy via "downloadTask"

ggerganov committed 1 year ago

Verified 4ed98b90

llama.swiftui : UX improvements

ggerganov committed 1 year ago

Verified 96294487

gitignore : xcode stuff

ggerganov committed 1 year ago

Verified d36ca171

comment #Preview & fix editorconfig check

jhen0409 committed 1 year ago

da44d452

update project.pbxproj

jhen0409 committed 1 year ago

a520e87e

add download buttons & expose llamaState.loadModel

jhen0409 committed 1 year ago

ce1df812

force to use n_gpu_layers on simulator

jhen0409 committed 1 year ago

ff87313d

llama.swiftui : initial bench functionality

ggerganov committed 1 year ago

Verified 6a868020

llama.swiftui : add bench button

ggerganov committed 1 year ago

Verified afd336f7

ggml : use ggml_row_size where possible (#4472)

slaren committed 1 year ago

Verified 6744dbe9

ggml : remove n_dims from ggml_tensor (#4469)

slaren committed 1 year ago

Verified cafcd4f8

py : add protobuf dependency (#4466)

strutive07 committed 1 year ago

Verified c50e4001

ggml : add ggml_row_size() (fixes llama out of space) (#4461)

LostRuins committed 1 year ago

Verified 20a68a70

ggml : fix OpenCL broadcast requirement for ggml_mul (close #4453)

ggerganov committed 1 year ago

Verified 55e87c37

convert : support loading vocab from fast tokenizer config (#3633)

strutive07 committed 1 year ago

Verified 873637af

readme : update supported model list (#4457)

BarfingLemurs committed 1 year ago

Verified 0353a184

server : fix handling of characters that span multiple tokens when streaming (#4446)

shibe2 committed 1 year ago

Verified 948ff137

sync : ggml (SD ops, tests, kernels) (#4444)

ggerganov committed 1 year ago

Verified 4d98d9a6

build : detect host compiler and cuda compiler separately (#4414)

cebtenzzre committed 1 year ago

Verified 70f806b8

common : add `--version` option to show build info in CLI (#4433)

yusiwen committed 1 year ago

Verified 9fb13f95

readme : update hot topics

ggerganov committed 1 year ago

Verified 113f9942

llama : add Mixtral support (#4406)

slaren committed 1 year ago

Verified 799a1cb1

server : tweak default sampling parameters (#4367)

kalomaze committed 1 year ago

Verified fecac456

english : use `typos` to fix comments and logs (#4354)

richardkiss committed 1 year ago

Verified 9494d7c4

build : target Windows 8 for standard mingw-w64 (#4405)

cebtenzzre committed 1 year ago

Verified 6138963f

llama : document logits_all deprecation (#4418)

crasm committed 1 year ago

Verified 6391817c

server : fix local model name in server (#4420)

epicfilemcnulty committed 1 year ago

Verified d9d4cfef

Older