About Janus Pro
Janus Pro is DeepSeek’s advanced multimodal AI model, offering enhanced text-to-image generation, improved training strategies, and scalable performance for industries like healthcare, e-commerce, and creative design.
About Janus Pro
The Next Evolution in Multimodal AI
Janus Pro is the latest iteration of its groundbreaking multimodal AI model, Janus. Building on the success of its predecessor, Janus Pro represents a significant leap forward in AI capabilities, offering enhanced performance, scalability, and versatility for text and image understanding and generation.
Enhanced Data Scaling
With access to larger and more diverse datasets, Janus Pro achieves deeper contextual understanding and generates more realistic, high-quality outputs.
Increased Model Size
The expanded model architecture allows Janus Pro to handle more complex tasks, making it ideal for applications requiring advanced multimodal capabilities.
Installation Guide for Janus Pro
Download Docker Desktop:
Visit the official Docker website and download the latest version of Docker Desktop for your operating system (Windows, macOS, or Linux).
Install Docker Desktop:
Run the installer and follow the on-screen instructions to complete the installation.
For Windows Users:
Install the Windows Subsystem for Linux (WSL) by running the following command in your terminal:
wsl --installRestart your computer if prompted.
Clone the Repository:
Open your terminal and run the following command to clone the Janus repository from GitHub:
git clone https://github.com/deepseek-ai/Janus.git
Navigate to the Project Directory:
Move into the cloned repository:
cd Janus
Open the Demo File:
Navigate to the
demofolder and open the fileapp_januspro.pyin your preferred code editor.
Change the Model Name:
Replace
deepseek-ai/Janus-Pro-7Bwithdeepseek-ai/Janus-Pro-1Bto load the lighter version of the model (4.1 GB), which is more suitable for local use.
Update the Launch Configuration:
Modify the last line of the script to ensure compatibility with Docker:
demo.queue(concurrency_count=1, max_size=10).launch( server_name="0.0.0.0", server_port=7860 )
Create a Dockerfile:
In the root directory of the project, create a file named
Dockerfileand add the following content:# Use the PyTorch base image FROM pytorch/pytorch:latest # Set the working directory inside the container WORKDIR /app # Copy the current directory into the container COPY . /app # Install necessary Python packages RUN pip install -e .[gradio] # Set the entrypoint for the container to launch your Gradio app CMD ["python", "demo/app_januspro.py"]
Build the Docker Image:
Run the following command in your terminal to build the Docker image:
docker build -t janus .
This process may take 10–15 minutes, depending on your internet speed.
Launch the Container:
Once the image is built, run the container using the following command:
docker run -p 7860:7860 janus
Access the Web Application:
Open your web browser and navigate to
http://localhost:7860to access the Janus Pro web application.
If you prefer to use the lighter Janus-Pro-1B model for faster performance on local machines, ensure the model name is updated in the app_januspro.py file.
