5fp commited on
Commit
382bd2f
·
verified ·
1 Parent(s): 13355db

Update Dockerfile

Browse files
Files changed (1) hide show
  1. Dockerfile +7 -6
Dockerfile CHANGED
@@ -4,7 +4,7 @@ ARG BASE_CPU_CONTAINER=ubuntu:${UBUNTU_VERSION}
4
  FROM ${BASE_CPU_CONTAINER} as build
5
 
6
  RUN apt-get update && \
7
- apt-get install -y build-essential git cmake wget
8
 
9
  WORKDIR /build
10
 
@@ -15,10 +15,11 @@ WORKDIR /build/llama.cpp
15
  # Ensure that LLAMA_CUBLAS is not set to 1
16
  ENV LLAMA_CUBLAS=0
17
 
18
- RUN mkdir build && \
19
- cd build && \
20
- cmake .. && \
21
- cmake --build . --config Release --target llama-server
 
22
 
23
  WORKDIR /data
24
  RUN wget https://huggingface.co/brunopio/Llama3-8B-1.58-100B-tokens-GGUF/resolve/main/Llama3-8B-1.58-100B-tokens-TQ2_0.gguf -nv -O model.gguf
@@ -28,7 +29,7 @@ FROM ${BASE_CPU_CONTAINER} as runtime
28
  WORKDIR /app
29
 
30
  # Copy the executable from the build stage
31
- COPY --from=build /build/llama.cpp/build/bin/llama-server /app
32
  COPY --from=build /data/model.gguf /data/model.gguf
33
  COPY ./run.sh /app/run.sh
34
  WORKDIR /app
 
4
  FROM ${BASE_CPU_CONTAINER} as build
5
 
6
  RUN apt-get update && \
7
+ apt-get install -y build-essential git make cmake wget
8
 
9
  WORKDIR /build
10
 
 
15
  # Ensure that LLAMA_CUBLAS is not set to 1
16
  ENV LLAMA_CUBLAS=0
17
 
18
+ RUN make llama-server
19
+ #RUN mkdir build && \
20
+ # cd build && \
21
+ # cmake .. && \
22
+ # cmake --build . --config Release --target llama-server
23
 
24
  WORKDIR /data
25
  RUN wget https://huggingface.co/brunopio/Llama3-8B-1.58-100B-tokens-GGUF/resolve/main/Llama3-8B-1.58-100B-tokens-TQ2_0.gguf -nv -O model.gguf
 
29
  WORKDIR /app
30
 
31
  # Copy the executable from the build stage
32
+ COPY --from=build /build/llama.cpp/llama-server /app
33
  COPY --from=build /data/model.gguf /data/model.gguf
34
  COPY ./run.sh /app/run.sh
35
  WORKDIR /app