ExternalVendorCode/lollms-webui
1
0
Fork 0
mirror of https://github.com/ParisNeo/lollms-webui.git synced 2024-12-18 20:17:50 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

updated

Browse Source
This commit is contained in:
Saifeddine ALOUI 2023-07-13 01:18:19 +02:00
parent e2012922ff
commit 703c34ed30
1 changed files with 34 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
docs/youtube/v3_installation.md Normal file
View File

@ -0,0 +1,34 @@
Hi there. i hope you like the new voice.
Just kidding. it is exactly the same but in another tone.
Today I'll show you how to install the new version of lord of large language models. I have changed the installation and run script a little bit to make it easy for every one and with no prior requirements on your PC.
The script installs miniconda and uses it to install all required tools, so you don't need to have python, git or even conda installed as was required by the previous versions. Even cuda is installed automatically on demand.
Now let's dive right in.
As usual, just go the the github page and press latest release.
If you notice I have changed the name of installers to be more explicit. take the installer that fits your needs, download it and run it.
Make sure the file is placed in a path that doesn't contain problematic naming like using long paths or so, as this can mess up conda.
Choose whether you want to use a GPU or not then proceed
Now let the installation take place. it will install all you need to build and run optimized and tailored versions of the bindings to your hardware.
Now you can run the win_run.bat script or linux_run.sh script to start the application.
As usual, select your personal folder path or sujt press enter to continue. The first time, the tool will download bindings library and personalities library.
Then you will be asked to select the first binding.
Let's select AutoGPTQ.
Now wait for it to be installed
Let's not install a model from here, just press back, and we will do the installation from the UI.
Let's go to the ui by browsing to http://localhost:9600. Notice that you can change this address if you want to share the service outside your pc, just call the run script with the option --host 0.0.0.0 or your ip address.
Let's search a model and install it.
Now let's select the installed model, apply changes, and save them.
Make sure you do save the changes as they would be lost if you don't.
Now you are ready for action.
I hope you like this video.
If you do, please consider leaving a comment, like and subscribe. you can also follow me on twitter @ParisNeo or come to my discord channel. Links are in the description.
Thanks for watching. See ya.