Skip to content

RVC Webui Premium 1-Click Installers for Windows, RunPod, Massed Compute, SimplePod with All GPUs Supporting Inlcuding RTX 1000, 2000, 3000, 4000, 5000 GPUs with fastest libraries #2779

@FurkanGozukara

Description

@FurkanGozukara

This is not an issue thread thank you

Get the app installer zip file from here : https://www.patreon.com/posts/149104996

Tutorial Video : https://youtu.be/ZRrzvD4wNys

Latest installer zip file : RVC_Installer.zip

  • This app was requested so I found the most up to date fork and forked and improved it
    • Using this app is not trivial so look for guides and tutorials
  • I have downloaded pre-trained audio models from here and tested and works
  • There are 450 GB of pre-trained voice models pth files in above repo that you can use
    • Put those pth files into Q:\RVC_Installer_V1\RVC_Premium\assets\weights
    • Put those index files into G:\RVC_Installer_V1\RVC_Premium\assets\indices
    • You will see auto downloaded models so you will understand how it works
  • Our installer will auto download all of the necessary models for voice training and other tasks automatically
  • Our installer will also download over 30 demo voices
  • I don't have much experience with this app but you can find lots of tutorials to learn more about it and this is the most up to date version forked from here : https://github.com/fumiama/Retrieval-based-Voice-Conversion-WebUI
  • Our installer will generate a Python 3.10 venv and install all the libraries inside it
    • Torch 2.9.1 with CUDA 13, Flash Attention, Sage Attention, xFormers, compatible with all GPUs starting from RTX 1000 to RTX 5000 GPUs
  • We also have installers for RunPod, Massed Compute and SimplePod
  • This is a very advanced version of original RVC Webui Premium developed by SECourses

10 April 2026 V4

  • Full tutorial video published that shows how to install and use on Windows, RunPod, SimplePod and Massed Compute : https://youtu.be/ZRrzvD4wNys
  • Upon request, merge voice feature implemented to generate custom mixed voices
  • Supports merging more than 2 voices with rates as well and including indexes (not mandatory)
  • Fairseq library installation was extremely hard on Windows therefore I hace pre-compiled it and now it is auto installed with support to all GPUs out ther
  • Download latest zip file, extract and overwrite all files and then Just run Windows_Install_and_Update.bat file

image

image

18 March 2026 V2

  • I have upgraded the app significantly
  • Now we are moved to newest Torch 2.9.1 and CUDA 13
  • Workin on all GPUs, Windows, Linux, RunPod, Massed Compute and SimplePod
  • Cloud installation tested and verified too
  • Now the installer will download 30+ pre-trained demo voices
  • Full preset save and load system implemented
  • Loop all models feature implemented thus you can loop and generate with all models and see how they work
  • Now all generations will be saved inside outputs folder with metadata
  • Now batch folder processing works perfect just give input and output folder path, it will process every file in given folder and save with same file name in output folder
  • Auto separate vocals + music and remix after conversion implemented thus with 1 click you can voice change full music audio files
  • Save as MP3 feature added
  • The parameters on the interface set to best as default
  • Uses below 4 GB VRAM and ultra fast
  • I have tested Windows_Start_Realtime_Voice_Changing_Desktop_GUI.bat and realtime voice change working too - taking speaking from your microphone and real time
  • For offline inference and training use Windows_Start_Voice_Change_And_Train_WebUI.bat
  • Command_To_Test_RunPod_SimplePod_GPU.txt added because out of 3 pods i rented, 2 had broken hardware so best to test
  • Make sure to select CUDA 12.9 and 13 filter in RunPod selection due to outdated NVIDIA drivers
  • Make a fresh installation with Windows_Install_and_Update.bat

image

screencapture-127-0-0-1-7865-2026-04-10-00_41_49

Windows Requirements

image

Massed Compute (Recommend Cloud) :

RunPod (Cloud):

App Screenshots

image

image

image

image

image

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions