ggerganov/llama.cpp

Pull Requests Commits

server: CI fix trailing space (#5728)

phymbert committed 1 year ago

Verified 4804215c

server: CI tests reduce build matrix (#5725)

phymbert committed 1 year ago

Verified 8a533f0d

llama : fix Gemma rope type (#5691)

ggerganov committed 1 year ago

Verified 269de86b

flake.lock: Update

github-actions[bot] committed 1 year ago

c3937339

server: tests - slow inference causes timeout on the CI (#5715)

phymbert committed 1 year ago

Verified e3965cf3

server: docs - refresh and tease a little bit more the http server (#5718)

phymbert committed 1 year ago

Verified 8b350356

llama : refactor k-shift implementation + KV defragmentation (#5691)

ggerganov committed 1 year ago

Verified bf08e006

server : fix crash when system prompt is bigger than batch size (#5714)

compilade committed 1 year ago

Verified f7625019

ggml-quants : provide ggml_vqtbl1q_u8 for 64bit compatibility (#5711)

rgryta committed 1 year ago

Verified abbabc5e

make : fix nvcc version is empty (#5713)

kwin1412 committed 1 year ago

Verified f1a98c52

readme : add Msty to UI list (#5618)

ashokgelal committed 1 year ago

Verified 7d548a18

server: logs - unified format and --log-format option (#5700)

phymbert committed 1 year ago

Verified 930b1780

server: concurrency fix + monitoring - add /metrics prometheus compatible endpoint (#5708)

phymbert committed 1 year ago

Verified d52d7819

cmake : fix compilation for Android armeabi-v7a (#5702)

rgryta committed 1 year ago

Verified 12894088

code : normalize enum names (#5697)

ggerganov committed 1 year ago

Verified ab336a9d

py : fix StableLM conversion after config.json changes (#5703)

aahouzi committed 1 year ago

Verified 69917dfa

server: continue to update other slots on embedding concurrent request (#5699)

phymbert committed 1 year ago

Verified 9e359a4f

IQ3_S: a much better alternative to Q3_K (#5676)

ikawrakow committed 1 year ago

Verified 4c4cb307

server: init functional tests (#5566)

phymbert committed 1 year ago

Verified 525213d2

server : add KV cache quantization options (#5684)

AlpinDale committed 1 year ago

Verified fd43d66f

convert : fix missing ftype for gemma (#5690)

cebtenzzre committed 1 year ago

Verified 54fbcd2c

mpt : do not duplicate token_embd.weight on disk (#5670)

cebtenzzre committed 1 year ago

Verified 15499eb9

gemma : use more bits for the token_embd.weight tensor (#5650)

ggerganov committed 1 year ago

Verified 96633eec

py : add Gemma conversion from HF models (#5647)

ggerganov committed 1 year ago

Verified 847eedbd

ggml : always define ggml_fp16_t as uint16_t (#5666)

ggerganov committed 1 year ago

Verified 7e4f339c

ggerganov committed 1 year ago

Verified 334f76fa

ggml : 32-bit arm compat (whisper/1891)

ggerganov committed 1 year ago

Verified efd56b1c

nix: init singularity and docker images (#5056)

SomeoneSerge committed 1 year ago

Verified 201294ae

py : minor fixes (#5668)

ggerganov committed 1 year ago

Verified 5a9e2f60

Add Gemma chat template (#5665)

ngxson committed 1 year ago

Verified 373ee3fb

Newer Older