Update app.py

app.py

@@ -1,14 +1,26 @@
-#@title # Setup
-%cd /content/
-!rm -rf llama.cpp
-!git clone https://github.com/ggerganov/llama.cpp.git
-%cd llama.cpp
-!make LLAMA_CUBLAS=1
+import subprocess
+import os
 
-# zephyr-7b-beta
-!wget https://huggingface.co/TheBloke/zephyr-7B-beta-GGUF/resolve/main/zephyr-7b-beta.Q6_K.gguf
+# Change directory to /content
+os.chdir("/content")
 
+# Remove the llama.cpp directory if it exists
+subprocess.run(["rm", "-rf", "llama.cpp"])
 
-#@title # inference
-%cd /content/llama.cpp
-!./server -m zephyr-7b-beta.Q6_K.gguf -ngl 9999 -c 0 --port 12345
+# Clone the llama.cpp repository
+subprocess.run(["git", "clone", "https://github.com/ggerganov/llama.cpp.git"])
+
+# Change directory to llama.cpp
+os.chdir("llama.cpp")
+
+# Build the project with LLAMA_CUBLAS=1
+subprocess.run(["make", "LLAMA_CUBLAS=1"])
+
+# Download the zephyr-7b-beta model
+subprocess.run(["wget", "https://huggingface.co/TheBloke/zephyr-7B-beta-GGUF/resolve/main/zephyr-7b-beta.Q6_K.gguf"])
+
+# Change directory to /content/llama.cpp
+os.chdir("/content/llama.cpp")
+
+# Run the server
+subprocess.run(["./server", "-m", "zephyr-7b-beta.Q6_K.gguf", "-ngl", "9999", "-c", "0", "--port", "12345"])