Spaces:
Paused
Paused
Commit
·
98911c0
1
Parent(s):
3ed2de1
Create Dockerfile
Browse files- Dockerfile +35 -0
Dockerfile
ADDED
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Grab a fresh copy of the Python image
|
2 |
+
FROM python:3.11-slim
|
3 |
+
|
4 |
+
# Install build and runtime dependencies
|
5 |
+
RUN apt-get update && \
|
6 |
+
apt-get install -y \
|
7 |
+
libopenblas-dev \
|
8 |
+
ninja-build \
|
9 |
+
build-essential \
|
10 |
+
pkg-config \
|
11 |
+
curl
|
12 |
+
|
13 |
+
RUN pip install -U pip setuptools wheel && \
|
14 |
+
CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" FORCE_CMAKE=1 pip install --verbose llama-cpp-python[server]
|
15 |
+
|
16 |
+
# Download model
|
17 |
+
RUN mkdir model && \
|
18 |
+
curl -L https://huggingface.co/TheBloke/OpenHermes-2.5-Mistral-7B-GGUF/resolve/main/openhermes-2.5-mistral-7b.Q4_K_M.gguf -o model/gguf-model.bin
|
19 |
+
|
20 |
+
COPY ./start_server.sh ./
|
21 |
+
COPY ./main.py ./
|
22 |
+
COPY ./index.html ./
|
23 |
+
|
24 |
+
# Make the server start script executable
|
25 |
+
RUN chmod +x ./start_server.sh
|
26 |
+
|
27 |
+
# Set environment variable for the host
|
28 |
+
ENV HOST=0.0.0.0
|
29 |
+
ENV PORT=7860
|
30 |
+
|
31 |
+
# Expose a port for the server
|
32 |
+
EXPOSE ${PORT}
|
33 |
+
|
34 |
+
# Run the server start script
|
35 |
+
CMD ["/bin/sh", "./start_server.sh"]
|