File size: 1,375 Bytes
78449ad
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b74b65c
78449ad
 
 
 
 
 
 
 
 
 
 
6498a05
78449ad
6498a05
 
 
78449ad
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
ARG UBUNTU_VERSION=22.04
ARG CUDA_VERSION=12.3.1
ARG BASE_CUDA_DEV_CONTAINER=nvidia/cuda:${CUDA_VERSION}-devel-ubuntu${UBUNTU_VERSION}
ARG BASE_CUDA_RUN_CONTAINER=nvidia/cuda:${CUDA_VERSION}-runtime-ubuntu${UBUNTU_VERSION}

FROM ${BASE_CUDA_DEV_CONTAINER} as build

ARG CUDA_DOCKER_ARCH=all

RUN apt-get update --fix-missing && \
    apt-get install -y --no-install-recommends git build-essential gcc cmake && \
    rm -rf /var/lib/apt/lists/*

WORKDIR /build

RUN git clone https://github.com/ggerganov/llama.cpp.git

WORKDIR /build/llama.cpp
#RUN git checkout fd5ea0f897ecb3659d6c269ef6f3d833e865ead7

ENV CUDA_DOCKER_ARCH=${CUDA_DOCKER_ARCH}
ENV LLAMA_CUBLAS=1

RUN mkdir build && \
    cd build && \
    cmake .. -DLLAMA_CUBLAS=ON && \
    cmake --build . --config Release

FROM ${BASE_CUDA_RUN_CONTAINER} as runtime
RUN apt-get update --fix-missing && \
    apt-get install -y --no-install-recommends wget libgomp1 python3 python3-pip && \
    rm -rf /var/lib/apt/lists/*
    
# Install huggingface_hub[cli]
RUN pip3 install huggingface_hub[cli]

WORKDIR /app

# Copy the executable from the build stage
COPY --from=build /build/llama.cpp/build/bin/server /app
COPY --from=build /build/llama.cpp/examples/server/public /app/public
COPY ./run.sh /app/run.sh
WORKDIR /app
EXPOSE 7867

# Make the script executable
RUN chmod +x run.sh

# CMD to run your script
CMD ./run.sh