VoltaML

Appearance

Docker ​

Docker is not recommended as it is updated less frequently than the Windows, WSL and Linux installations. However, it is the easiest way to get started with VoltaML in the cloud or on Vast.ai.

WARNING

This setup will require WSL2 on Windows. If you installed WSL or Docker before it was released, you will need to switch manually to WSL2.

This is the easiest setup possible that should just work. Docker provides reproducible environments and makes it easy to run the application on any system.

Requirements ​

  • Operating system: Windows or Linux
  • Graphics card: NVIDIA GPU with CUDA support
  • Graphics card for AITemplate: RTX 40xx, RTX 30xx, H100, A100, A10, A30, V100, T4
  • Driver version: 515+ with CUDA
  • Docker: Docker Desktop

Main Branch (default - stable) ​

1. Clone the repository ​

bash
git clone https://github.com/VoltaML/voltaML-fast-stable-diffusion --single-branch

2. Get inside the directory ​

bash
cd voltaML-fast-stable-diffusion

3. Edit the docker-compose.yml file ​

yaml
version: "3.7"

services:
  voltaml:
    image: stax124/volta:latest
    pull_policy: always 
    environment:
      - HUGGINGFACE_TOKEN=${HUGGINGFACE_TOKEN}
      - LOG_LEVEL=${LOG_LEVEL}
      - FASTAPI_ANALYTICS_KEY=${FASTAPI_ANALYTICS_KEY}
      - DISCORD_BOT_TOKEN=${DISCORD_BOT_TOKEN}
      - EXTRA_ARGS=${EXTRA_ARGS}
    volumes: 
      - XXX:/app/data # XXX is the path to the folder where all the outputs will be saved 
      - YYY/.cache/huggingface:/root/.cache/huggingface # YYY is path to your home folder (you may need to change the YYY/. cache/huggingface to YYY\.cache\huggingface on Windows) 
    ports:
      - "5003:5003"
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              capabilities: ["gpu"]

volumes:
  cache: {}
  • XXX is the path to the directory where you want to store all the data (converted models, outputs)
  • YYY is the path to your home directory (C:\Users\YOUR_USERNAME or /home/USER).
  • pull_policy: always means that you will always try to get the latest image. Feel free to set it to missing to download only if it isn't present on your system

4. Edit the .env file ​

bash
# Hugging Face Token (https://huggingface.co/settings/tokens)
HUGGINGFACE_TOKEN=YOUR_TOKEN_HERE 

# Logging level (DEBUG, INFO, WARNING, ERROR, CRITICAL)
LOG_LEVEL=INFO

# [Optional] Analytics (https://my-api-analytics.vercel.app/generate) (https://my-api-analytics.vercel.app/dashboard)
FASTAPI_ANALYTICS_KEY=

# [Optional] Discord Bot Token (https://discord.com/developers/applications)
DISCORD_BOT_TOKEN=

# [Optional] Extra arguments for the API
EXTRA_ARGS=

5. Run the container ​

bash
docker compose run --service-ports voltaml

or if you have older version of docker

bash
docker-compose run -p 5003:5003 voltaml

Experimental Branch (development - unstable) ​

If you would like to have the latest features and bug fixes, and you do not mind having unstable container, then feel free to use this branch.

1. Clone the repository ​

bash
git clone https://github.com/VoltaML/voltaML-fast-stable-diffusion -b experimental --single-branch

2. Get inside the directory ​

bash
cd voltaML-fast-stable-diffusion

3. Edit the docker-compose.yml file ​

yaml
..
image: stax124/volta:experimental
..
volumes:
  - XXX:/app/data
  - YYY/.cache/huggingface:/root/.cache/huggingface
..

Where XXX is the path to the directory where you want to store all the data (converted models, outputs) and YYY is the path to your home directory (C:\Users\YOUR_USERNAME or /home/USER).

Make sure you replace the image property with stax124/volta:experimental

4. Edit the .env file ​

bash
HUGGINGFACE_TOKEN=PLACE_YOUR_TOKEN_HERE
LOG_LEVEL=INFO # INFO, DEBUG, WARNING, ERROR

5. Run the container ​

bash
docker compose run --service-ports voltaml

or if you have older version of docker

bash
docker-compose run -p 5003:5003 voltaml