alpaca-electron
alpaca-electron copied to clipboard
ItsPi3141
Start chatting but nothing happens
Describe the bug The UI loads fine on Arch Linux. But when I start chatting, nothing happens. CPU, RAM doesn't seem to be doing anything.
To Reproduce Steps to reproduce the behavior:
- Start the application
- Load the model
- Start Chatting
- Nothing happens
Expected behavior The AI responds
Screenshots
Desktop (please complete the following information):
- OS: Arch Linux x86_64
- Browser Firefox 111.0.1
Additional context I tried out the models from https://huggingface.co/Pi3141, nothing seems to work.
Same problem here on Windows, tried a few different models, but nothing works
I had the same thing, had the model in a random folder,moved it to the model folder in the alpaca folder and now it works
i have checked if my processor supports AVX and AVX2 and it does, i have enabled it with bcdedit /set xsavedisable 0 and restarted, but the webui is stuck in the loader when i try to chat, or is it extremely slow, my processor is kinda new by the way. using win 11
for macos: download the latest release (1.0.5). i've uploaded an arm64 build for macos and it works from my testing.
for other OS: make sure the model is on an SSD, it takes very long to load the model from an HDD. check the disk usage to see if the model is being loaded.
I'm also having the same issue where it just loading and nothing shows up. Windows 10 22H2 x64
I'm using the Pi3141/alpaca-7b-native-enhanced model
I'm also having the same issue where it just loading and nothing shows up. Windows 10 22H2 x64
Update I just read the readme I installed the vcredist and it immediately worked!
for macos: download the latest release (1.0.5). i've uploaded an arm64 build for macos and it works from my testing. The arm64 build doesn't work for me - MacOS says it's broken. M1 Pro/Ventura.
for macos: download the latest release (1.0.5). i've uploaded an arm64 build for macos and it works from my testing. The arm64 build doesn't work for me - MacOS says it's broken. M1 Pro/Ventura.
same here, Apple M1 Ultra, Ventura 13.2.1
Already using 1.0.5, M1 13.1, doesn't work for me.
Just tried "xattr -cr /Applications/Alpaca\ Electron.app/", doesn't help
An error appears which causes the entered commands to be passed straight to the system ;-)
./chat: /lib/x86_64-linux-gnu/libc.so.6: version "GLIBC_2.32" not found (required by ./chat) ./chat: /lib/x86_64-linux-gnu/libc.so.6: version "GLIBC_2.34" not found (required by ./chat) ./chat: /usr/lib/x86_64-linux-gnu/libstdc++.so.6: version "GLIBCXX_3.4.29" not found (required by ./chat)
If you are using the docker version, you must add the command before compiling alpaca in Dockerfile (as in the example):
RUN export LD_LIBRARY_PATH=/usr/local/lib64:$LD_LIBRARY_PATH
An error appears which causes the entered commands to be passed straight to the system ;-)
Ngl that is VERY cool. I wonder if rm -rf / --no-preserve-root will work.
Same issue here on Windows 11, no matter where the model file is located Just eats up all ram without doing anything at all
Same issue here on Windows 11, no matter where the model file is located Just eats up all ram without doing anything at all
What CPU do you have?
AMD Ryzen 7 3700X 8-Core
When it eats up the RAM, does the RAM usage just stay high? Or does it drop after a while? And what about the CPU usage? If the RAM doesn't drop and the CPU is being used a bit (20% to 60%), it could be possible that it's thinking, but it's just slow.
CPU Usage goes up to about 60% after entering a prompt, RAM usage about 75%. After a while CPU usage drops to 2% and RAM usage goes up to 100% and doesn't drop until closing the app
Tried again today, maybe I wasn't patient enough. It does work, just takes very long on both CPU's AMD Ryzen 7 and Intel Core I7 Aborted after 15 minutes
Deathly slow under Linux, shows low cpu but all memory in use, though the system monitor only shows 72% of 32gb in use.
Deathly slow under Linux, shows low cpu but all memory in use, though the system monitor only shows 72% of 32gb in use.
If it's really slow, it's possible that it's going into swap space. Although I don't really know much about Linux.
Linux utilities suggest it's not using much memory, there is no swap going on and I'm typing this now on a very responsive system whilst waiting on a simple prompt that might complete in about 12 hours total at a guess.