Server Deployment

Dockerfile

@@ -8,4 +8,4 @@ RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
 
 COPY . .
 
-CMD ["gunicorn", "-b", "0.0.0.0:7860","model:app"]
+CMD ["uvicorn", "-b", "0.0.0.0:7860","model:app"]

main.py

@@ -3,7 +3,7 @@ from llama_cpp import Llama
 from huggingface_hub import hf_hub_download
 
 # Initialize the Llama model with chat format set to "llama-2"
-llm = Llama(model_path="E:/langchain-chat-gui-main/langchain-chat-gui-main/llama-2-7b-chat.Q2_K.gguf", chat_format="llama-2")
+llm = Llama(model_path="./llama-2-7b-chat.Q2_K.gguf", chat_format="llama-2")
 
 # Define the system prompt
 system_prompt = (