ggml-org/llama.cpp

Pull Requests Commits

ci : add cublas to windows release

Green-Sky committed 2 years ago

Verified 31ff9e2e

examples : read chat prompts from a template file (#1196)

khimaros committed 2 years ago

Verified 6daa09d8

minor : fix whitespaces (#1302)

ggerganov committed 2 years ago

Verified bca9ad93

minor : fix trailing whitespaces

ggerganov committed 2 years ago

Verified e2a937ca

scripts : platform independent script to verify sha256 checksums (#1203)

KASR committed 2 years ago

Verified b0c71c7b

examples : various prompt and example fixes (#1298)

CRD716 committed 2 years ago

Verified a8a2efdc

llama : only copy used KV cache in get / set state (#1272)

ejones committed 2 years ago

Verified e216aa04

Process escape sequences given in prompts (#1173)

DannyDaemonic committed 2 years ago

Verified 2485d7a4

Handle signals properly on Windows (#1123)

DannyDaemonic committed 2 years ago

Verified 13b0c68e

Call sh on build-info.sh (#1294)

DannyDaemonic committed 2 years ago

Verified 55bc5f09

fix build-info.h for git submodules (#1289)

kuvaus committed 2 years ago

Verified 9daff419

fix missing parameters in `llama_init_from_gpt_params` (#1293)

slaren committed 2 years ago

Verified bf4b22ff

examples : add llama_init_from_gpt_params() common function (#1290)

deadprogram committed 2 years ago

Verified 67c77799

llama : fix compile warnings

ggerganov committed 2 years ago

Verified 0e6cbff1

ggml : fix 32-bit ARM

ggerganov committed 2 years ago

Verified 5d5817ca

examples : improve vertical alignment of a few variables (#1286)

deadprogram committed 2 years ago

Verified 8c9be35f

ggml : fix ppc64le build error and make cmake detect Power processors (#1284)

mgiessing committed 2 years ago

Verified cc0bb723

llama : allow 0 as a seed number. (#1275)

rbrisita committed 2 years ago

Verified 2bb992f0

main : switch input_noecho to input_echo to remove negation (#979)

deadprogram committed 2 years ago

Verified e2cd5069

ggml: add names to tensors (#1268)

slaren committed 2 years ago

Verified 2d099e51

Add git-based build information for better issue tracking (#1232)

DannyDaemonic committed 2 years ago

Verified f4cef87e

cuBLAS: refactor and optimize f16 mat mul performance (#1259)

slaren committed 2 years ago

Verified 58b367c2

llama : update stubs for systems without mmap and mlock (#1266)

xloem committed 2 years ago

Verified ea3a0ad6

ggml : fix ggml_used_mem() (#1264)

KerfuffleV2 committed 2 years ago

Verified 2bdc0964

llama : fix session load / save (#1263)

ggerganov committed 2 years ago

Verified 70269cae

cuBLAS: fall back to pageable memory if pinned alloc fails (#1233)

slaren committed 2 years ago

Verified b925f1f1

llama : let context be const when accessing const data (#1261)

grencez committed 2 years ago

Verified 90b19bd6

ggml : fix UB (int << 31)

ggerganov committed 2 years ago

Verified 7ff0dcd3

build: add armv{6,7,8} support to cmake (#1251)

prusnak committed 2 years ago

Verified 6f796992

common : better default number of threads (#934)

jon-chuang committed 2 years ago

Verified a5d30b1f

Older