ggerganov/llama.cpp

Pull Requests Commits
Server: Use multi-task for embeddings endpoint (#6001)
ngxson committed 1 year ago
Verified 99b71c06
ci : remove tidy-review (#6021)
slaren committed 1 year ago
Verified 306d34be
ggml : reuse quantum structs across backends (#5943)
ggerganov committed 1 year ago
Verified 8030da7a
ggml : fix UB in IQ2_S and IQ3_S (#6012)
ggerganov committed 1 year ago
Verified 184215e7
sycl : update IQ1_S kernels (WIP - not working!) (#5995)
ggerganov committed 1 year ago
Verified 48358b2e
grammar : fix unnecessarily retained pointer to rules (#6003)
gliptic committed 1 year ago
Verified 5cdb3717
1.5 bit: we can do even better (#5999)
ikawrakow committed 1 year ago
Verified 44ca159f
llama : more consistent names of count variables (#5994)
ggerganov committed 1 year ago
Verified 05b06210
llama : refactor unicode stuff (#5992)
ggerganov committed 1 year ago
Verified 83796e62
Update server docker image URLs (#5997)
GetTuh committed 1 year ago
Verified 828defef
Server: format error to json (#5961)
ngxson committed 1 year ago
Verified caa106d4
ggml, ci : Windows ARM runner and build fixes (#5979)
Xarbirus committed 1 year ago
Verified 3202361c
server : maintain chat completion id for streaming responses (#5988)
mscheong01 committed 1 year ago
Verified 332bdfd7
cmake : fix subdir for `LLAMA_METAL_EMBED_LIBRARY` (#5985)
giladgd committed 1 year ago
Verified ecab1c75
llama : fix F16/F32 downcast + improve names (#5980)
ggerganov committed 1 year ago
Verified ee35600b
Better 1.5 bit quantization (#5971)
ikawrakow committed 1 year ago
Verified be858f62
[SYCL] Add q3_s and q1_s (#5886)
abhilash1910 committed 1 year ago
Verified ef3ced26
[SYCL] Add support for SYCL Nvidia target (#5738)
AidanBeltonS committed 1 year ago
Verified 3814a073
metal : move mm_id indices to shared mem (#5982)
ggerganov committed 1 year ago
Verified bb6d00bb
android : fix utf8 decoding error (#5935)
zhangfuwen committed 1 year ago
Verified 7ab7b733
readme : update hot topics
ggerganov committed 1 year ago
Verified d9f65c97
sync : ggml
ggerganov committed 1 year ago
Verified b838b53a
ggml : try fix 32-bit arm compat (whisper/1938)
ggerganov committed 1 year ago
Verified df4dc3e7
ggml : remove __constant__ specifier for CUDA tables (#5940)
ggerganov committed 1 year ago
Verified bf47a5ee
server: ci: windows build and tests (#5968)
phymbert committed 1 year ago
Verified fa8a809a
llama : add support for GritLM (#5959)
dranger003 committed 1 year ago
Verified bcebd7db
grammar : verify parsed state (#5950)
HanClinto committed 1 year ago
Verified 2960eae8
nix: update flake.lock (#5969)
ggerganov committed 1 year ago
Verified c7854147
server: benchmark: chat/completions scenario and other llm servers comparison (#5941)
phymbert committed 1 year ago
Verified 621e86b3
server : print chat template info
ggerganov committed 1 year ago
Verified 77d1ac7e
Terms Of Use Privacy Policy Refund Policy Impressum