Update app.py

app.py

@@ -15,7 +15,7 @@ app = gr.load(
     ]
 ).launch()
 """
-
+"""
 # Pipeline
 
 import gradio as gr
@@ -39,4 +39,57 @@ app = gr.Interface(
     examples = [
         ["Hello, World."]
     ]
-).launch()
+).launch()
+"""
+
+import gradio as gr
+from huggingface_hub import InferenceClient
+import os
+
+hf_token = os.getenv("HF_TOKEN")
+client = InferenceClient(api_key=hf_token)
+
+def chatbot(input_text, history):
+    # Prepare the conversation messages
+    messages = [{"role": "user", "content": input_text}]
+
+    # Add conversation history (if exists)
+    if history:
+        for user_input, bot_response in history:
+            messages.append({"role": "user", "content": user_input})
+            messages.append({"role": "assistant", "content": bot_response})
+
+    # Generate model response
+    stream = client.chat.completions.create(
+        model="google/gemma-2-2b-it", 
+        messages=messages, 
+        temperature=0.5,
+        max_tokens=2048,
+        top_p=0.7,
+        stream=True
+    )
+
+    # Collect response from the model
+    bot_response = ""
+    for chunk in stream:
+        bot_response += chunk.choices[0].delta.content
+
+    # Update the conversation history
+    history.append((input_text, bot_response))
+    return bot_response, history
+
+# Create Gradio Interface
+with gr.Blocks() as demo:
+    chatbot_ui = gr.Chatbot(label="Gemma Chatbot").style(height=400)
+    text_input = gr.Textbox(show_label=False, placeholder="Type your message here...")
+    state = gr.State([])  # Keeps track of conversation history
+
+    def user_input_handler(user_input, chatbot_ui, state):
+        bot_response, state = chatbot(user_input, state)
+        chatbot_ui.append((user_input, bot_response))
+        return chatbot_ui, state
+
+    text_input.submit(user_input_handler, [text_input, chatbot_ui, state], [chatbot_ui, state])
+
+# Run the application
+demo.launch()