Docker image for LLaVA: Large Language and Vision Assistant

Important

If you are using the 13b model, CUDA will result in OOM errors with a GPU that has less than 48GB of VRAM, so A6000 or higher is recommended.

Installs

• Ubuntu 22.04 LTS
• CUDA 11.8
• Python 3.10.12
• LLaVA v1.1.1
• Torch 2.1.0
• llava-v1.5-7b model

You can add an environment variable called MODEL to use the 13b model (llava-v1.5-13b) instead of the default 7b version. The 7b version is now the default so that it can be used on GPUs with less VRAM available.

Available on RunPod

This image is designed to work on RunPod. You can use my custom RunPod template to launch it on RunPod.

Running Locally

Install Nvidia CUDA Driver

• Linux
• Windows

Start the Docker container

docker run -d \
  --gpus all \
  -v /workspace \
  -p 3000:3001 \
  -p 8888:8888 \
  -e JUPYTER_PASSWORD=Jup1t3R! \
  ashleykza/llava:1.1.2

You can obviously substitute the image name and tag with your own.

Acknowledgements

1. Matthew Berman for giving me a demo on LLaVA, as well as his amazing YouTube videos.

Community and Contributing

Pull requests and issues on GitHub are welcome. Bug fixes and new features are encouraged.

You can contact me and get help with deploying your container to RunPod on the RunPod Discord Server below, my username is ashleyk.

Appreciate my work?