Update app.py

app.py

@@ -23,8 +23,13 @@ import pprint
 import io
 import torch
 
+from huggingface_hub import InferenceApi
+
 
 try:
+    # Set long session timeout for space.
+    inference = InferenceApi(repo_id="MVPilgrim/SemanticSearch", timeout=1800)
+    
     #############################################
     # Logging setup including weaviate logging. #
     #############################################
@@ -348,8 +353,8 @@ try:
     # Initialize the LLM. #
     #######################
     #model_path = "/app/llama-2-7b-chat.Q4_0.gguf"
-    #model_path = "/app/Llama-3.2-3B-Instruct-Q4_0.gguf"
-    model_path = "Meta-Llama-3.1-8B-Instruct-Q8_0.gguf"
+    model_path = "/app/Llama-3.2-3B-Instruct-Q4_0.gguf"
+    #model_path = "Meta-Llama-3.1-8B-Instruct-Q8_0.gguf"
     if 'llm' not in st.session_state:
         logger.info("### Initializing LLM.")
         llm = Llama(model_path,