Update app.py

app.py

@@ -3,7 +3,6 @@ from huggingface_hub import InferenceClient
 import nltk
 import PyPDF2
 
-# Download required NLTK resources
 nltk.download("punkt", quiet=True)
 nltk.download("punkt_tab", quiet=True)
 
@@ -14,12 +13,7 @@ def respond_chunked(message, history, system_message, max_tokens, temperature, t
     combined_response = ""
     for chunk in chunks:
         chunked_system_message = f"{system_message}\n\nFile Content Chunk:\n{chunk}"
-        messages = [{"role": "system", "content": chunked_system_message}]
-        for user, assistant in history:
-            if user:
-                messages.append({"role": "user", "content": user})
-            if assistant:
-                messages.append({"role": "assistant", "content": assistant})
+        messages = [{"role": "system", "content": chunked_system_message}] + history
         messages.append({"role": "user", "content": message})
         try:
             completion = client.chat_completion(
@@ -104,9 +98,11 @@ with gr.Blocks() as demo:
     def chat_function(user_message, history, file_content, system_prompt, max_tokens, temperature, top_p):
         if not user_message.strip():
             return "", history
-        history.append((user_message, ""))
+        # Get response from the model with chunking
         assistant_response = respond_chunked(user_message, history, system_prompt, max_tokens, temperature, top_p, file_content)
-        history[-1] = (user_message, assistant_response)
+        # Append user and assistant messages in the correct format
+        history.append({"role": "user", "content": user_message})
+        history.append({"role": "assistant", "content": assistant_response})
         return "", history
 
     send_button = gr.Button("Send")
@@ -122,4 +118,4 @@ with gr.Blocks() as demo:
         outputs=[user_input, chatbot]
     )
 
-    demo.launch(server_name="0.0.0.0", server_port=7860)
+    demo.launch(server_name="0.0.0.0", server_port=7860, share=True)