Closed Temtaime closed 8 months ago
It happened before when I did not have enough RAM for the model to run. Which model do you try to run? How many GB of memory do you have? What is the CPU of your computer?
I have 64gb
ram, tried wizard-vicuna-30b-uncensored-q4_k_s
, cpu is ryzen 5900x
.
Do other models work for you?
Try this:
# cd to catai directory
cd `where catai | rev | cut -c10- | rev`/lib/node_modules/catai/
# update node-llama-cpp to the latest version
npm i node-llama-cpp@latest
If it also does not work, we can try to recompile it:
npx node-llama-cpp download
After that tell me if it works, I want to check the issue more deeply :)
:tada: This issue has been resolved in version 3.0.1 :tada:
The release is available on:
Your semantic-release bot :package::rocket:
Hello
npm i node-llama-cpp@latest
executed without errors, but it still was not working
After recompiling with npx node-llama-cpp download
it started to work, thanks !
Not working on Ubuntu 22.04.3 LTS and node 18
On it, I think the cpp binaries for AMD do not work properly
:tada: This issue has been resolved in version 3.1.0-beta.1 :tada:
The release is available on:
Your semantic-release bot :package::rocket:
You can fix this issue, by using catai beta and rebuilding the binaries
npm i -g catai@beta
catai cpp
Please refer to the troubleshooting before opening an issue. You might find the solution there.
Describe the bug Followed instructions, when
catai up
is used it only opens a page for 1 time, then process exits.Screenshots
Desktop (please complete the following information):
catai --version
) 3.0.0node --version
) 20