Update app.py

app.py

@@ -1,21 +1,32 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
 
+# Initialize an empty list to store the conversation history
+conversation_history = []
+
 # Function to interact with the model using the Inference API
 def chat_with_model(user_input, hf_api_key):
+    global conversation_history
+    
     # Initialize the InferenceClient with the provided API key
     client = InferenceClient(api_key=hf_api_key)
 
-    # Define the messages for the chat (system message tailored for a code assistant)
-    messages = [
-        {"role": "system", "content": "You are a code assistant that helps with code generation, debugging, and explanations."},
-        {"role": "user", "content": user_input}
-    ]
+    # Add the user's message to the conversation history
+    conversation_history.append({"role": "user", "content": user_input})
+
+    # Define the system message (defining the assistant role)
+    system_message = {
+        "role": "system", 
+        "content": "You are a code assistant that helps with code generation, debugging, and explanations."
+    }
+
+    # Add system message to the conversation history
+    conversation_history.insert(0, system_message)
 
     # Create a stream for chat completions using the API
     stream = client.chat.completions.create(
         model="Qwen/Qwen2.5-Coder-32B-Instruct", 
-        messages=messages, 
+        messages=conversation_history, 
         max_tokens=500,
         stream=True
     )
@@ -24,6 +35,10 @@ def chat_with_model(user_input, hf_api_key):
     response = ""
     for chunk in stream:
         response += chunk.choices[0].delta.content
+    
+    # Add the assistant's response to the conversation history
+    conversation_history.append({"role": "assistant", "content": response})
+    
     return response
 
 # Create the Gradio interface
@@ -39,4 +54,4 @@ iface = gr.Interface(
 )
 
 # Launch the interface
-iface.launch()
+iface.launch()