llm download

Dockerfile

@@ -16,9 +16,10 @@ RUN pip install torch
 
 RUN pip install accelerate
 
-RUN CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install 'llama-cpp-python[server]' --upgrade --force-reinstall --no-cache-dir
-RUN python3 -m llama_cpp.server --model ./models/7B/llama-model.gguf
+# Install hugging face hub to download llama2 model
+RUN pip install --upgrade huggingface_hub
 
+RUN CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install 'llama-cpp-python[server]' --upgrade --force-reinstall --no-cache-dir 
 # Install requirements.txt 
 RUN pip install --no-cache-dir --upgrade -r /requirements.txt
 

question_paper.py

@@ -8,8 +8,13 @@ from transformers import AutoTokenizer
 
 from llama_cpp import Llama
 
+from huggingface_hub import hf_hub_download
+
 # Load the model
 
+hf_hub_download(repo_id="bevangelista/Llama-2-7b-chat-hf-GGUF-Q4_K_M", filename="Llama-2-7b-chat-hf-GGUF-Q4_K_M.gguf")
+
+
 app = FastAPI()
 @app.get("/")
 def llama():