MandaChain Studios

AI Video Restoration & Enhancement Setup Guide

🚀 Choose Your Deployment

Cloud

☁️ Deploy on RunPod

Best for: No local GPU, fast setup, on-demand usage

Setup Time ~5 minutes
GPU Required Rented (RTX 4090 recommended)
Cost Pay-per-use credits
1

Create a RunPod Account

Sign up at runpod.io and add credits to your account.

2

Launch a GPU Pod

Select an RTX 4090 (the 4090 template will work with 3090 as well) pod (or any NVIDIA GPU with CUDA 12.1+ (except Blackwell 50xx or newer)).

  • Click "Change Template" and search "MandaChain"
  • Click the appropriate template for the GPU series:
    mandachain/video-restoration-4090:v1
    or
    mandachain/video-restoration-5090:v1
  • After selecting the template, under "Storage Configuration" select Volume disk and increase the size to 30GB
  • Click deploy on-demand
  • The pod will download the template and model
  • The template will take ~5 minutes to be ready
  • Click the logs tab to see the status.
3

Connect & Open the UI

Once the pod starts, click Connect, then click the Interface link. The interface will load automatically — upload your video and start restoring.

Note: The app checks for CUDA 12.1+ on startup. If you see a "node is too old" warning, switch to a pod with a newer driver (we've tested numerous pods and very rarely find a pod that is outdated this much).
Local

🖥️ Deploy Locally

Best for: Own NVIDIA GPU, privacy, unlimited usage

Setup Time ~15 minutes
GPU Required RTX 30/40-series (10 GB+ VRAM)
Cost Free (your hardware)
1

Install Docker Desktop

Download and install Docker Desktop. On Windows, make sure WSL 2 backend is enabled (Docker will prompt you during setup).

2

Install the NVIDIA Container Toolkit

Docker needs GPU access. Install the latest NVIDIA Container Toolkit and make sure your NVIDIA driver is up to date (CUDA 12.1+ required). On Windows with Docker Desktop + WSL 2, GPU pass-through works out of the box with a recent driver — no extra toolkit install needed inside WSL.

3

Pull the Docker Image

Open a terminal and pull the pre-built image: (the 4090 image will work on 30/40-series GPUs, the 5090 image is for 50-series GPUs)

docker pull mandachain/video-restoration-4090:v1
or
docker pull mandachain/video-restoration-5090:v1

4

Run the Container

Start the container with GPU access and port mapping:

docker run --gpus all -p 7860:7860 mandachain/video-restoration-4090:v1
or
docker run --gpus all -p 7860:7860 mandachain/video-restoration-5090:v1

To keep restored videos accessible on your host machine, mount an output folder:

docker run --gpus all -p 7860:7860 -v C:\Videos\restored:/app/outputs mandachain/video-restoration-4090:v1
or
docker run --gpus all -p 7860:7860 -v C:\Videos\restored:/app/outputs mandachain/video-restoration-5090:v1

5

Open the UI

Once the container is running and Gradio has loaded, open your browser to http://localhost:7860. Upload your video and start restoring.

GPU Compatibility: The video-restoration-4090:v1 image has been tested on RTX 4090 via RunPod but should work with any NVIDIA RTX 30-series or 40-series GPU (e.g. RTX 3080, 3090, 4070, 4080). You need at least ~8 GB VRAM for 720p and ~12 GB+ for 1080p output. Performance will vary by card.
Troubleshooting: If you see a CUDA version error on startup, update your NVIDIA driver to the latest version. If Docker cannot see your GPU, verify it with docker run --gpus all nvidia/cuda:12.1.0-base-ubuntu22.04 nvidia-smi.