ggml-org/llama.cpp

Pull Requests Commits

plamo : testing

ggerganov committed 2 years ago

Verified 3c734f49

okdshin committed 2 years ago

9339ffc9

remove develop code

okdshin committed 2 years ago

907b9218

update kqv code

okdshin committed 2 years ago

febc6359

okdshin committed 2 years ago

ca8f6986

okdshin committed 2 years ago

f76fd392

okdshin committed 2 years ago

86d5348f

fix norm_rms_eps hparam

okdshin committed 2 years ago

a22040a8

able to compile

okdshin committed 2 years ago

4a3ef4f2

okdshin committed 2 years ago

9d492365

okdshin committed 2 years ago

b2330f57

add tensor loading

okdshin committed 2 years ago

4c585b4c

okdshin committed 2 years ago

feb0966a

decode : fix logits_valid for legacy API (#4516)

cebtenzzre committed 2 years ago

Verified 2994f0c5

readme : update hot topics

ggerganov committed 2 years ago

Verified b1306c43

llama.swiftui : add bench functionality (#4483)

ggerganov committed 2 years ago

Verified 800a489e

gguf-py : fail fast on nonsensical special token IDs (#4489)

cebtenzzre committed 2 years ago

Verified f7f468a9

build : Check the ROCm installation location (#4485)

Su3h7aM committed 2 years ago

Verified 919c4066

finetune : keep allocs alive until all allocations are done (#4486)

slaren committed 2 years ago

Verified 45668633

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

Older