Update app.py

app.py

@@ -1,33 +1,21 @@
 import gradio as gr
-import random
 from huggingface_hub import InferenceClient
-
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
 def respond(message, history):
-
-    messages = [{"role": "system", "content": "You are a friendly chatbot."}]
-    if history: 
+    messages = [{"role": "system", "content": "You are a friendly chatbot.""}]
+    if history:
         messages.extend(history)
-
-    messages.append({"role" : "user", "content" : message})
-
-    response = client.chat_completion(
-        messages, 
-        max_tokens = 30,
-        temperature =.9,
-        top_p=.7
-
-)
-
-    
-    return response["choices"][0]["message"]["content"].strip()
-
-def echo(messaged, history):
-    my_list = ["sure", "ask again", "maybe"]
-    return random.choice(my_list)
-chatbot = gr.ChatInterface(respond, type = "messages")
-
-
-
+    messages.append({"role": "user", "content": message})
+    response = ""
+    for message in client.chat_completion(
+        messages,
+        max_tokens=500,
+        temperature=0.2,
+        top_p=0.9,
+        stream=True
+    ):
+        token = message.choices[0].delta.content
+        response += token
+        yield response
+chatbot = gr.ChatInterface(respond, type=“messages”)
 chatbot.launch()