> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bunkerbio.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Run bunker-fold in Docker

> Use the official Bunker Longevity Docker image. Pre-built with bunker-fold, CUDA runtime, and Python 3.11 for GPU-backed local development and CI.

The official Bunker Longevity Docker image ships with `bunker-fold`, a CUDA runtime, and Python 3.11 pre-installed. Use it for GPU-backed local development, reproducible CI, or as a base for your own containers.

## Pull the image

```bash theme={null}
docker pull ghcr.io/bunker-longevity/bunker-fold:latest
```

## Run a prediction

```bash theme={null}
docker run --rm --gpus all \
  -e BUNKER_API_KEY \
  ghcr.io/bunker-longevity/bunker-fold:latest \
  python -c "from bunker_fold import Bunker; \
             print(Bunker().models.bunkerfold_3.predict(sequence='MVLSPADK').confidence)"
```

## Interactive shell

```bash theme={null}
docker run --rm -it --gpus all \
  -e BUNKER_API_KEY \
  ghcr.io/bunker-longevity/bunker-fold:latest \
  bash
```

## Build on top of the image

```dockerfile theme={null}
FROM ghcr.io/bunker-longevity/bunker-fold:latest

WORKDIR /app
COPY . .
RUN pip install -r requirements.txt

CMD ["python", "run.py"]
```

## Pin a version

```bash theme={null}
docker pull ghcr.io/bunker-longevity/bunker-fold:1.4.2
```

## Verify

```bash theme={null}
docker run --rm ghcr.io/bunker-longevity/bunker-fold:latest \
  python -c "import bunker_fold; print(bunker_fold.__version__)"
```