Running an rtx3060 with 12GBvram - managed to get this model working on method in link in description

#28
by planetfrog - opened

https://youtu.be/ByV5w1ES38A

hope this helps someone. I just followed the instructions, then moved this model / files to the new models folder.

Start-webui.bat file:

@echo off

@echo Starting the web UI...

cd /D "%~dp0"

set MAMBA_ROOT_PREFIX=%cd%\installer_files\mamba
set INSTALL_ENV_DIR=%cd%\installer_files\env

if not exist "%MAMBA_ROOT_PREFIX%\condabin\micromamba.bat" (
call "%MAMBA_ROOT_PREFIX%\micromamba.exe" shell hook >nul 2>&1
)
call "%MAMBA_ROOT_PREFIX%\condabin\micromamba.bat" activate "%INSTALL_ENV_DIR%" || ( echo MicroMamba hook not found. && goto end )
cd text-generation-webui

call python server.py --auto-devices --cai-chat --threads 8 --wbits 4 --groupsize 128

:end
pause

Also adjusted the HDD virtual memory setting to be managed by applications

Hmmm I just downloaded this model and loaded it with the oobabooga UI - no changes to any settings / configuration file. I see 9GB memory used while loading the model and it goes upto 11.2 GB during inference. (same GPU)

Wow that was a huge help, been struggling to get a working environment !

You can easily run any models using this method.

Sign up or log in to comment