Multimodal-Large-Language-Model (MLLM)

Thank you for checking out the Multimodal-Large-Language-Model project. Please note that this project was created for research purposes.

For a more robust and well-developed solution, you may consider using open-webui/open-webui with ollama/ollama.

Documentation

You can access the project documentation at [GitHub Pages].

Host requirements

• Docker: [Installation Guide]
• Docker Compose: [Installation Guide]
• Compatibile with Linux and Windows Host
• Ensure port 8501 and 11434 are not already in use
• You should have at least 8 GB of RAM available to run the 7B models, 16 GB to run the 13B models, and 32 GB to run the 33B models. [Source]
• Project can be ran on either CPU or GPU

Running on GPU

• NVIDIA Container Toolkit (Linux) [Installation Guide]
• NVIDIA CUDA Toolkit (Windows) [Installation]
• WSL (Windows) [Installation]

Tested Model(s)

Model Name	Size	Link
llava:7b	4.7GB	Link
llava:34b	20GB	Link

Llava is pulled and loaded by default, other models from Ollama can be added into ollama/ollama-build.sh

Usage

Note

Project will run on GPU by default. To run on CPU, use the docker-compose.cpu.yml instead

1. Clone this repository and navigate to project folder

git clone https://github.com/NotYuSheng/Multimodal-Large-Language-Model.git
cd Multimodal-Large-Language-Model

2. Build the Docker images:

docker compose build

3. Run images

docker compose up -d

4. Access Streamlit webpage from host

<host-ip>:8501

API calls to Ollama server can be made to

<host-ip>:11434

Star History

Thank you for showing your support by starring this project. Your recognition is greatly appreciated! ⭐