diff --git a/dockerfiles/Dockerfile.vllm b/dockerfiles/Dockerfile.vllm new file mode 100644 index 0000000..d90b9f2 --- /dev/null +++ b/dockerfiles/Dockerfile.vllm @@ -0,0 +1,25 @@ +# Use vLLM's vllm-openai server image as the base +FROM vllm/vllm-openai:v0.6.3.post1 + +# Define a build argument for the working directory, defaulting to /workspace +ARG WORKDIR_ARG=/workspace + +# Set the working directory +WORKDIR ${WORKDIR_ARG} + +# Install necessary build dependencies for sentencepiece +RUN apt-get update && apt-get install -y \ + pkg-config \ + cmake \ + build-essential + +# Copy functionary code and requirements into workspace +COPY . . + +# Install additional Python dependencies +RUN --mount=type=cache,target=/root/.cache/pip \ + python3 -m pip install .[vllm] + +# Override the VLLM entrypoint with the functionary server +ENTRYPOINT ["python3", "server_vllm.py", "--model", "meetkai/functionary-small-v3.2", "--host", "0.0.0.0", "--max-model-len", "8192"] +CMD []