Update app.py

app.py

@@ -73,9 +73,6 @@ global_data['models'] = model_manager.load_all_models()
 
 class ChatRequest(BaseModel):
     message: str
-    top_k: int = 50
-    top_p: float = 0.95
-    temperature: float = 0.7
 
 def normalize_input(input_text):
     return input_text.strip()
@@ -94,9 +91,9 @@ def remove_duplicates(text):
     return '\n'.join(unique_lines)
 
 @GPU(duration=0)
-def generate_model_response(model, inputs, top_k, top_p, temperature):
+def generate_model_response(model, inputs):
     try:
-        response = model(inputs, top_k=top_k, top_p=top_p, temperature=temperature)
+        response = model(inputs)
         return remove_duplicates(response['choices'][0]['text'])
     except Exception as e:
         print(f"Error generating model response: {e}")
@@ -108,7 +105,7 @@ async def generate(request: ChatRequest):
         inputs = normalize_input(request.message)
         with ThreadPoolExecutor() as executor:
             futures = [
-                executor.submit(generate_model_response, model, inputs, request.top_k, request.top_p, request.temperature)
+                executor.submit(generate_model_response, model, inputs)
                 for model in global_data['models'].values()
             ]
             responses = [{'model': model_name, 'response': future.result()} for model_name, future in zip(global_data['models'].keys(), as_completed(futures))]