llama-cpp-server-13b

Sleeping

muryshev commited on Dec 18, 2023

Commit

fe8c3af

1 Parent(s): c84c1b2

Cuda disabled

Files changed (1) hide show

Dockerfile CHANGED Viewed

@@ -21,11 +21,12 @@ ENV LLAMA_CUBLAS=1
 RUN mkdir build && \
     cd build && \
-    cmake .. -DLLAMA_CUBLAS=ON && \
     cmake --build . --config Release
 WORKDIR /data
-RUN wget https://huggingface.co/IlyaGusev/saiga2_7b_gguf/resolve/main/model-q8_0.gguf -O model.gguf
 FROM ${BASE_CUDA_RUN_CONTAINER} as runtime

 RUN mkdir build && \
     cd build && \
+#    cmake .. #-DLLAMA_CUBLAS=ON && \
+    cmake .. && \
     cmake --build . --config Release
 WORKDIR /data
+RUN wget https://huggingface.co/IlyaGusev/saiga2_7b_gguf/resolve/main/model-q8_0.gguf -nv -O model.gguf
 FROM ${BASE_CUDA_RUN_CONTAINER} as runtime