Commit History

fix layout, max size back to 1, llama.cpp doesn't like parallel calls
80c7d2e

winglian commited on

loading torch doesn't play nice, just use config for now
54d2e53

winglian commited on

try to fix combining gr.interface with blocks, try to increase concurrency on larger gpus
dce6894

winglian commited on

support tabbed interface and fix instruct to use GPU
ec779d5

winglian commited on

patience
58f7b09

winglian commited on

update copy
a68164e

winglian commited on

more fixes to config file based llama
776e6fd

winglian commited on

update instruct to use gpu, add verbose description
3143561

winglian commited on

default to instruct for now
2ade250

winglian commited on

cleanup instruct app. add queue so UI notifies users
ed16925

winglian commited on

single concurrency is probably safest
2b019dd

winglian commited on

base model in config is unnecessary, and fix grammar
946b2e7

winglian commited on

move byline around
d90dc27

winglian commited on

tweak the system prompt
496b4f8

winglian commited on

add prediction settings to ui
ce7dde7

winglian commited on

docker cp doesn't support wildcards
988bc04

winglian commited on

link model attributions, use config.yml for some of the chat settings, increase context size
1dc6c65

winglian commited on

rm docker implementation, add llama-cpp-python builder github actions, update copy to identify model in ui
e3ba05b

winglian commited on

fix chat history, update settings to use GPU
4cc03d2

winglian commited on

require torch and nvidia-cublas
0a981aa

winglian commited on

rollback to gradio
51c11ce

winglian commited on

try out docker sdk
6f7e42d

winglian commited on

Update requirements.txt
29b3b7f

winglian commited on

let's see what happens on a GPU
39f7680

winglian commited on

retry on failed hub fetch
de2057e

winglian commited on

initial commit for ggml instruct
7b590e5

winglian commited on

initial commit
85686d8

winglian commited on