Update app.py

app.py

@@ -5,20 +5,20 @@ from ctransformers import AutoModelForCausalLM
 # Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
 llm = AutoModelForCausalLM.from_pretrained("TheBloke/WizardLM-7B-uncensored-GGUF", model_file="WizardLM-7B-uncensored.Q4_K_M.gguf", model_type="llama", stream=True)
 
-previous_responses = []
 
 def generate_response(message):
+    
     for text in llm(message):
         yield text
+    yi
 
 def chatbot(message, history):
-    global previous_responses
     response_generator = generate_response(message)
-    response_list = list(response_generator)
-    previous_responses += response_list
-    for response in response_list:
-        time.sleep(0.3)  # Optional delay for a natural chat feel
-        yield response
+    for response in response_generator:
+        time.sleep(0.3)
+        # Optional delay for a natural chat feel
+        x= "".join(response)
+        yield x
 
 iface = gr.ChatInterface(chatbot)
-iface.launch()
+iface.launch()