Update app.py

app.py

@@ -36,6 +36,7 @@ def load_quantized_model(model_id, model_basename):
 model, tokenizer = load_quantized_model(model_name_or_path, "model.safetensors")
 
 
+
 def load_model_norm():
     if torch.cuda.is_available():
         print("CUDA is available. GPU will be used.")
@@ -51,6 +52,7 @@ def load_model_norm():
 
     return model, tokenizer 
 
+    
 # Function to generate a response using the model
 def generate_response(prompt: str) -> str:
     PERSONA_NAME = "Ivana"
@@ -94,6 +96,7 @@ async def api_home():
     return {'detail': 'Welcome to Eren Bot!'}
 
     
+    
 # Endpoint to start a new conversation thread
 @app.post('/api/start_conversation')
 async def start_conversation(request: Request):
@@ -110,6 +113,7 @@ async def start_conversation(request: Request):
     return {'thread_id': thread_id, 'response': response}
 
 
+    
 # Endpoint to get the response of a conversation thread
 @app.get('/api/get_response/{thread_id}')
 async def get_response(thread_id: int):
@@ -124,9 +128,11 @@ async def get_response(thread_id: int):
 
     return {'response': response}
 
+
 @app.post('/api/chat')
 async def chat(request: Request, chat_input: ChatInput):
-    prompt = chat_input.prompt
+    data = await request.json()
+    prompt = data.get('prompt')
 
     # Generate a response based on the prompt
     response = generate_response(prompt)