ggml-org/llama.cpp

Pull Requests Commits

llama : allow overriding GGUF metadata when loading model (#4092)

KerfuffleV2 committed 1 year ago

Verified 5aa365d8

sampling : custom samplers order (#4285)

MaggotHATE committed 1 year ago

Verified 52c8bc3c

swift : revert compiler checks for swift package (#4332)

kchro3 committed 1 year ago

Verified e4b76bbe

simple : update error message for KV cache check (#4324)

danbev committed 1 year ago

Verified 23b5e12e

swift : fix concatenation method to avoid invalid UTF8 stringfication (#4325)

ensan-hcl committed 1 year ago

Verified d208995c

swift : fix prompt tokenization logic (#4321)

ensan-hcl committed 1 year ago

Verified 5c9f90cb

grammar-parser : fix typo (#4318)

eltociear committed 1 year ago

Verified 4fa44e84

ggml : reuse ggml_get_n_tasks() in ggml_graph_plan() (#4308)

ggerganov committed 1 year ago

Verified fbbc4282

ggml : fix soft max out-of-bounds access (#4307)

ggerganov committed 1 year ago

Verified adf3de4f

server : fix OpenAI API `stop` field to be optional (#4299)

Mardak committed 1 year ago

Verified 33e171d1

py : add grammar to oai like api (#4294)

neph1 committed 1 year ago

Verified 6949b50d

llama : pad KV cache size (#4280)

ggerganov committed 1 year ago

Verified d7b800b8

llama : avoid using "optional" keyword (#4283)

ggerganov committed 1 year ago

Verified 5a7d3125

llama : support optional tensors (#4283)

ggerganov committed 1 year ago

Verified d5a1cbde

swift : fix token_to_piece implementation (#4278)

ensan-hcl committed 1 year ago

Verified b220222a

build : enable libstdc++ assertions for debug builds (#4275)

cebtenzzre committed 1 year ago

Verified 511f52c3

llama : support attention bias on LLaMA architecture (#4283)

RealJosephus committed 1 year ago

Verified 03562f3a

llama : add Qwen support (#4281)

simonJJJ committed 1 year ago

Verified 37c746d6

llama : fix integer overflow during quantization (#4284)

ggerganov committed 1 year ago

Verified 880f5797

py : add requirements file for convert-hf-to-gguf.py (#4277)

danbev committed 1 year ago

Verified 8d6d9f03

ggml : add ggml_soft_max_ext (#4256)

ggerganov committed 1 year ago

Verified ef47ec18

server : add --log-disable to disable logging to file (#4260)

ziedbha committed 1 year ago

Verified 1d144112

server : add single-client multi-prompt support (#4232)

ziedbha committed 1 year ago

Verified f43f0936

make : fix Apple clang determination bug (#4272)

WillCorticesAI committed 1 year ago

Verified d2809a3b

build : fix build info generation and cleanup Makefile (#3920)

cebtenzzre committed 1 year ago

Verified 15f5d960

llava : ShareGPT4V compatibility (vision encoder only loading) (#4172)

cmp-nct committed 1 year ago

Verified 33c9892a

main : pass LOG_TEE callback to llama.cpp log (#4033)

AndrewGodfrey committed 1 year ago

Verified 8efa0f6e

readme : fix (#4135)

vodkaslime committed 1 year ago

Verified 524907aa

docker : add finetune option (#4211)

jooray committed 1 year ago

Verified 3bd2c7ce

batched.swift : update README.md (#4214)

ensan-hcl committed 1 year ago

Verified bde629bb

Newer Older