diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000..140e743
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,54 @@
+# This Dockerfile uses the nvidia cuda container but will work without cuda too
+
+FROM nvidia/cuda:12.1.0-devel-ubuntu20.04
+ENV CUDAARCHS='60'
+ENV TRIMESH_VERSION='2020.03.04'
+ENV CMAKE_VERSION='3.20.4'
+
+WORKDIR /installation/
+
+RUN apt update
+RUN apt install -y --no-install-recommends apt-utils
+RUN apt install -y build-essential libglm-dev libgomp1 git mesa-common-dev libglu1-mesa-dev libxi-dev wget ninja-build
+
+WORKDIR /installation/cuda_voxelizer
+
+COPY . .
+
+RUN wget -q -O ./cmake-install.sh https://github.com/Kitware/CMake/releases/download/v${CMAKE_VERSION}/cmake-${CMAKE_VERSION}-Linux-x86_64.sh 
+RUN chmod u+x ./cmake-install.sh
+RUN mkdir /cmake
+RUN ./cmake-install.sh --skip-license --prefix=/cmake
+
+WORKDIR /installation/
+RUN git clone --single-branch --depth 1 -b ${TRIMESH_VERSION} https://github.com/Forceflow/trimesh2.git trimesh2
+WORKDIR /installation/trimesh2
+RUN pwd
+RUN make all -j $(nproc)
+
+ENV PATH="${PATH}:$HOME"/cmake/bin
+
+WORKDIR /installation/cuda_voxelizer
+
+RUN cmake -GNinja \
+        -DTrimesh2_INCLUDE_DIR="/installation/trimesh2/include" \
+        -DTrimesh2_LINK_DIR="/installation/trimesh2/lib.Linux64" \
+        -S . -B ./build
+
+
+RUN PATH=$PATH:"$HOME"/cmake/bin
+RUN cmake --build ./build --parallel $(nproc)
+
+FROM nvidia/cuda:12.1.0-devel-ubuntu20.04
+
+WORKDIR /app/
+
+COPY --from=0 /installation/cuda_voxelizer/build/cuda_voxelizer ./
+COPY --from=0 /installation/cuda_voxelizer/test_models/bunny.OBJ ./
+
+RUN ./cuda_voxelizer -o binvox -thrust -f ./bunny.OBJ -s 256
+#RUN ./cuda_voxelizer -o binvox -thrust -f ./bunny.OBJ -s 512
+#RUN ./cuda_voxelizer -o binvox -thrust -f ./bunny.OBJ -s 1024
+#RUN ./cuda_voxelizer -o binvox -thrust -f ./bunny.OBJ -s 2048
+
+ENTRYPOINT ["/app/cuda_voxelizer"]
diff --git a/README.md b/README.md
index 1c2bd06..3590c65 100644
--- a/README.md
+++ b/README.md
@@ -40,6 +40,22 @@ The project has the following build dependencies:
  * [GLM](http://glm.g-truc.net/0.9.8/index.html) for vector math. Any recent version will do.
  * [OpenMP](https://www.openmp.org/) for multi-threading.
 
+### Docker Compose
+1. Install [Docker](https://docs.docker.com/get-docker/) & [Docker Compose](https://docs.docker.com/compose/)
+2. Install the [NVIDIA Container Toolkit](https://github.com/NVIDIA/nvidia-docker)
+
+Then you can use the app like this:
+```bash
+docker compose run cuda_voxelizer cuda_voxelizer -o binvox -thrust -f ./test_models/bunny.OBJ -s 256
+
+```
+This should create a binvox file in the test_models folder. 
+If you have a cuda enabled gpu it will use it(maybe you have to lower the cuda arch variable in the Dockerfile).
+You can always add more folders inside of the docker-compose.yml.
+
+With this way, you dont have to install all the dependencies yourself.
+In the following ways you have to do the steps described in the Dockerfile yourself.
+
 ### Build using CMake (Windows, Linux)
 
 After installing dependencies, do `mkdir build` and `cd build`, followed by:
diff --git a/docker-compose.yml b/docker-compose.yml
new file mode 100644
index 0000000..9cd268e
--- /dev/null
+++ b/docker-compose.yml
@@ -0,0 +1,7 @@
+version: '3'
+services:
+  cuda_voxelizer:
+    image: forceflow/cuda_voxelizer 
+    build: .
+    volumes:
+      - ./test_models/:/app/test_models/