Update app.py

app.py

@@ -14,7 +14,8 @@ load_dotenv()
 llm = HuggingFaceInferenceAPI(
     # model_name="meta-llama/Meta-Llama-3-8B-Instruct",
     # model_name="mistralai/Mixtral-8x7B-Instruct-v0.1",
-    model_name="nvidia/Llama3-ChatQA-1.5-70B",
+    # model_name="nvidia/Llama3-ChatQA-1.5-70B", ## 144GB > 10GB
+    model_name= "Nexusflow/NexusRaven-V2-13B",
     token=os.getenv("TOKEN"),
     # model_config['protected_namespaces'] = ()
 )