Update app.py

app.py

@@ -4,20 +4,20 @@ 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)
-list = []
+history = []
+
 def generate_response(message):
+    global history
     for text in llm(message):
-        toe = list.join(text)
-        yield toe
-    
+        response = ''.join(text)
+        history.append(response)
+        yield ' '.join(history)
 
 def chatbot(message, history):
     response_generator = generate_response(message)
     for response in response_generator:
-        time.sleep(0.3)
-        # Optional delay for a natural chat feel
-        
-        yield response 
+        time.sleep(0.3)  # Optional delay for a natural chat feel
+        yield response
 
 iface = gr.ChatInterface(chatbot)
-iface.launch()
+iface.launch()