ggml-org/llama.cpp

Pull Requests Commits

server : disable llm logs if SERVER_VERBOSE is off (#3792)

olexiyb committed 2 years ago

Verified 0ffc92d2

server : fix grammar being ignored (#4494)

AdithyanI committed 2 years ago

Verified 8edd2b40

server : fix possible ambiguity in content type charset (#4501)

z80maniac committed 2 years ago

Verified eb16dae7

server : allow requests larger than 8K (#4500)

mzcu committed 2 years ago

Verified 62bd52b7

Link to cublas dynamically on Windows even with LLAMA_STATIC (#4506)

bullno1 committed 2 years ago

Verified 5daa5f54

lora : add support for non-llama models (#3333)

slaren committed 2 years ago

Verified c6c4fc08

llama : sanity checks for access to logits (#4274)

cebtenzzre committed 2 years ago

Verified 8a5be3bd

server : add optional API Key Authentication example (#4441)

ShadovvBeast committed 2 years ago

Verified 88ae8952

ggml : group mul_mat_id rows by matrix (cpu only) (#4480)

slaren committed 2 years ago

Verified ee4725a6

ggml : use ggml_row_size where possible (#4472)

slaren committed 2 years ago

Verified 6744dbe9

ggml : remove n_dims from ggml_tensor (#4469)

slaren committed 2 years ago

Verified cafcd4f8

py : add protobuf dependency (#4466)

strutive07 committed 2 years ago

Verified c50e4001

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

LostRuins committed 2 years ago

Verified 20a68a70

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

ggerganov committed 2 years ago

Verified 55e87c37

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

strutive07 committed 2 years ago

Verified 873637af

readme : update supported model list (#4457)

BarfingLemurs committed 2 years ago

Verified 0353a184

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

shibe2 committed 2 years ago

Verified 948ff137

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

ggerganov committed 2 years ago

Verified 4d98d9a6

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

cebtenzzre committed 2 years ago

Verified 70f806b8

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

yusiwen committed 2 years ago

Verified 9fb13f95

readme : update hot topics

ggerganov committed 2 years ago

Verified 113f9942

llama : add Mixtral support (#4406)

slaren committed 2 years ago

Verified 799a1cb1

server : tweak default sampling parameters (#4367)

kalomaze committed 2 years ago

Verified fecac456

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

richardkiss committed 2 years ago

Verified 9494d7c4

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

cebtenzzre committed 2 years ago

Verified 6138963f

llama : document logits_all deprecation (#4418)

crasm committed 2 years ago

Verified 6391817c

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

epicfilemcnulty committed 2 years ago

Verified d9d4cfef

ggml : increased GGML_MAX_PARAMS to allow finetuning of 70b models (#4424)

Taikono-Himazin committed 2 years ago

Verified 41a11aaf

Update README.md (#4388)

y10ab1 committed 2 years ago

Verified 8a7b2fa5

grammar : revert the replacement of llama_token_to_piece with id_to_token (#4396)

AlienKevin committed 2 years ago

Verified e18f7345

Newer Older