ggml-org/llama.cpp

Pull Requests Commits
metal : build metallib + fix embed path (#6015)
ggerganov committed 2 years ago
Verified 381da2d9
embedding : print cosine similarity (#899)
ggerganov committed 2 years ago
Verified 0fd6c1f0
readme : update details about running llama in Termux on Android (#6039)
wanix1988 committed 2 years ago
Verified 19885d20
readme : update API changes and hot topics
ggerganov committed 2 years ago
Verified 76a936c8
grammar : handle missing "root" node (#6004)
HanClinto committed 2 years ago
Verified 46362837
llama : add pipeline parallelism support (#6017)
slaren committed 2 years ago
Verified f30ea47a
test-backend-ops : skip CPU backend by default (#6028)
slaren committed 2 years ago
Verified d8fd0ccf
Update get version (#6025)
AidanBeltonS committed 2 years ago
Verified b3d97860
Server: Use multi-task for embeddings endpoint (#6001)
ngxson committed 2 years ago
Verified 99b71c06
ci : remove tidy-review (#6021)
slaren committed 2 years ago
Verified 306d34be
ggml : reuse quantum structs across backends (#5943)
ggerganov committed 2 years ago
Verified 8030da7a
ggml : fix UB in IQ2_S and IQ3_S (#6012)
ggerganov committed 2 years ago
Verified 184215e7
sycl : update IQ1_S kernels (WIP - not working!) (#5995)
ggerganov committed 2 years ago
Verified 48358b2e
grammar : fix unnecessarily retained pointer to rules (#6003)
gliptic committed 2 years ago
Verified 5cdb3717
1.5 bit: we can do even better (#5999)
ikawrakow committed 2 years ago
Verified 44ca159f
llama : more consistent names of count variables (#5994)
ggerganov committed 2 years ago
Verified 05b06210
llama : refactor unicode stuff (#5992)
ggerganov committed 2 years ago
Verified 83796e62
Update server docker image URLs (#5997)
GetTuh committed 2 years ago
Verified 828defef
Server: format error to json (#5961)
ngxson committed 2 years ago
Verified caa106d4
ggml, ci : Windows ARM runner and build fixes (#5979)
Xarbirus committed 2 years ago
Verified 3202361c
server : maintain chat completion id for streaming responses (#5988)
mscheong01 committed 2 years ago
Verified 332bdfd7
cmake : fix subdir for `LLAMA_METAL_EMBED_LIBRARY` (#5985)
giladgd committed 2 years ago
Verified ecab1c75
llama : fix F16/F32 downcast + improve names (#5980)
ggerganov committed 2 years ago
Verified ee35600b
Better 1.5 bit quantization (#5971)
ikawrakow committed 2 years ago
Verified be858f62
[SYCL] Add q3_s and q1_s (#5886)
abhilash1910 committed 2 years ago
Verified ef3ced26
[SYCL] Add support for SYCL Nvidia target (#5738)
AidanBeltonS committed 2 years ago
Verified 3814a073
metal : move mm_id indices to shared mem (#5982)
ggerganov committed 2 years ago
Verified bb6d00bb
android : fix utf8 decoding error (#5935)
zhangfuwen committed 2 years ago
Verified 7ab7b733
readme : update hot topics
ggerganov committed 2 years ago
Verified d9f65c97
sync : ggml
ggerganov committed 2 years ago
Verified b838b53a
Terms Of Use Privacy Policy Refund Policy Impressum