ztime commited on
Commit
e8482a8
1 Parent(s): f5f8f18

change port export

Browse files
Files changed (2) hide show
  1. Dockerfile +2 -1
  2. start_server.sh +1 -1
Dockerfile CHANGED
@@ -16,6 +16,7 @@ RUN apt-get update && \
16
 
17
  RUN git clone https://github.com/ggerganov/llama.cpp.git llamacpp --depth 1 && \
18
  cd llamacpp && \
 
19
  cmake -B build -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS && \
20
  cmake --build build --config Release --target main server && \
21
  cp build/bin/* ../ && \
@@ -42,4 +43,4 @@ ENV PORT=7860
42
  EXPOSE ${PORT}
43
 
44
  # Run the server start script
45
- CMD ["/bin/sh", "./start_server.sh"]
 
16
 
17
  RUN git clone https://github.com/ggerganov/llama.cpp.git llamacpp --depth 1 && \
18
  cd llamacpp && \
19
+ sed -i 's/v1\/chat/api\/v1\/chat/g' examples/server/server.cpp && \
20
  cmake -B build -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS && \
21
  cmake --build build --config Release --target main server && \
22
  cp build/bin/* ../ && \
 
43
  EXPOSE ${PORT}
44
 
45
  # Run the server start script
46
+ CMD ["/bin/sh", "./start_server.sh"]
start_server.sh CHANGED
@@ -3,7 +3,7 @@
3
  # For mlock support
4
  ulimit -l unlimited
5
 
6
- /main -m model/gguf-model.bin --color -p '你好,你是谁' -n 2 -c 2
7
 
8
  /server --host 0.0.0.0 --port 7860 -m model/gguf-model.bin
9
  /llamacpp/build/bin/server --host 0.0.0.0 --port 7860 -m model/gguf-model.bin
 
3
  # For mlock support
4
  ulimit -l unlimited
5
 
6
+ /main -m model/gguf-model.bin --color -p '你好,你是谁' -n 64 -c 64
7
 
8
  /server --host 0.0.0.0 --port 7860 -m model/gguf-model.bin
9
  /llamacpp/build/bin/server --host 0.0.0.0 --port 7860 -m model/gguf-model.bin