add vllm dockerfile
This commit is contained in:
parent
e0af2ea9c5
commit
e3c57d874a
|
@ -0,0 +1 @@
|
|||
`docker run --shm-size 14g --gpus all -v /storage/models/awq/MythoMax-L2-13B-AWQ:/models/MythoMax-L2-13B-AWQ -e ENV_API_SERVER_ARGS="--model /models/MythoMax-L2-13B-AWQ --quantization awq --host 0.0.0.0 --port 7000 --max-num-batched-tokens 8192 --gpu-memory-utilization 1" -d cyberes_vllm_cloud`
|
|
@ -0,0 +1,45 @@
|
|||
FROM nvidia/cuda:11.8.0-devel-ubuntu22.04 as build
|
||||
|
||||
RUN apt-get update && apt-get install -y git python3-pip python3-venv wget unzip && rm -rf /var/lib/apt/lists/*
|
||||
|
||||
RUN git clone https://git.evulid.cc/cyberes/local-llm-server.git /local-llm-server
|
||||
|
||||
WORKDIR /local-llm-server
|
||||
|
||||
RUN python3 -m venv /venv
|
||||
RUN /venv/bin/pip install -r requirements.txt
|
||||
RUN /venv/bin/pip install git+https://github.com/vllm-project/vllm
|
||||
|
||||
ADD "https://www.random.org/cgi-bin/randbyte?nbytes=10&format=h" skipcache
|
||||
|
||||
RUN git fetch; git reset --hard origin/master
|
||||
|
||||
RUN mkdir -p /app
|
||||
COPY ./rathole-client.toml /app/client.toml
|
||||
RUN wget https://github.com/rapiz1/rathole/releases/download/v0.4.8/rathole-x86_64-unknown-linux-gnu.zip -O /tmp/rathole.zip
|
||||
RUN unzip -j /tmp/rathole.zip -d /tmp
|
||||
RUN rm /tmp/rathole.zip
|
||||
RUN cp /tmp/rathole /app
|
||||
|
||||
FROM nvidia/cuda:11.8.0-base-ubuntu22.04 as runtime
|
||||
|
||||
RUN apt-get update && apt-get install -y supervisor && rm -rf /var/lib/apt/lists/*
|
||||
|
||||
COPY --from=build /local-llm-server /local-llm-server
|
||||
COPY --from=build /venv /venv
|
||||
COPY --from=build /app /app
|
||||
RUN cp /local-llm-server/other/vllm/Docker/supervisord.conf /etc/supervisor/conf.d/supervisord.conf
|
||||
|
||||
RUN useradd -ms /bin/bash apiserver
|
||||
RUN chown -R apiserver:apiserver /local-llm-server
|
||||
|
||||
RUN apt-get update && apt-get install -y python3 python3-pip wget aria2
|
||||
RUN pip3 install --upgrade pip setuptools wheel
|
||||
RUN pip install jupyterlab
|
||||
|
||||
EXPOSE 7000
|
||||
EXPOSE 8888 # jupyter
|
||||
CMD ["bash", "-c", "source /etc/bash.bashrc && /usr/bin/supervisord"]
|
||||
|
||||
# To test your config, uncomment this and comment out the other CMD.
|
||||
#CMD ["/app/rathole", "-c", "/app/client.toml"]
|
|
@ -0,0 +1,27 @@
|
|||
[supervisord]
|
||||
nodaemon=true
|
||||
|
||||
[program:api_server]
|
||||
command=/venv/bin/python /local-llm-server/other/vllm/vllm_api_server.py --host 0.0.0.0 --port 7000 %(ENV_API_SERVER_ARGS)s
|
||||
autostart=true
|
||||
autorestart=true
|
||||
stdout_logfile=/dev/fd/1
|
||||
stdout_logfile_maxbytes=0
|
||||
stderr_logfile=/dev/fd/2
|
||||
stderr_logfile_maxbytes=0
|
||||
user=apiserver
|
||||
environment=HOME="/home/apiserver",USER="apiserver"
|
||||
|
||||
[program:proxy]
|
||||
command=/app/rathole -c /app/client.toml
|
||||
autostart=true
|
||||
autorestart=true
|
||||
stdout_logfile=/dev/fd/1
|
||||
stdout_logfile_maxbytes=0
|
||||
stderr_logfile=/dev/fd/2
|
||||
stderr_logfile_maxbytes=0
|
||||
user=apiserver
|
||||
environment=HOME="/home/apiserver",USER="apiserver"
|
||||
|
||||
[program:jupyter]
|
||||
command=bash -c "source /etc/bash.bashrcPIP_DISABLE_PIP_VERSION_CHECK=1 jupyter lab --allow-root --ip=0.0.0.0 --no-browser --ServerApp.trust_xheaders=True --ServerApp.disable_check_xsrf=False --ServerApp.allow_remote_access=True --ServerApp.allow_origin='*' --ServerApp.allow_credentials=True"
|
Reference in New Issue