Update app.py

app.py

@@ -12,7 +12,8 @@ warnings.filterwarnings("ignore")
 logging.set_verbosity_error()
 
 # Global variables
-MODEL_ID = "microsoft/bitnet-b1.58-2B-4T"
+# Updated to use a model that's actually available on Hugging Face
+MODEL_ID = "microsoft/phi-2"  # Alternative: "microsoft/phi-1_5" or any other available model
 MAX_LENGTH = 2048
 MAX_NEW_TOKENS = 512
 TEMPERATURE = 0.7
@@ -37,7 +38,7 @@ def load_model_and_tokenizer():
         tokenizer = AutoTokenizer.from_pretrained(
             MODEL_ID,
             use_fast=True,
-            trust_remote_code=True  # Added to trust remote code
+            trust_remote_code=True
         )
         
         # Load model with optimizations for limited resources
@@ -46,7 +47,7 @@ def load_model_and_tokenizer():
             device_map="auto",
             torch_dtype=torch.bfloat16,
             load_in_4bit=True,
-            trust_remote_code=True  # Added to trust remote code
+            trust_remote_code=True
         )
         
         print("Model and tokenizer loaded successfully!")
@@ -162,7 +163,7 @@ def health_check():
 # Gradio Web UI
 def create_ui():
     with gr.Blocks() as demo:
-        gr.Markdown("# BitNet Specialist Chatbot with Step-by-Step Thinking")
+        gr.Markdown("# AI Assistant with Step-by-Step Thinking")
         
         with gr.Row():
             with gr.Column():
@@ -218,8 +219,4 @@ if __name__ == "__main__":
     
     # Create and launch Gradio interface
     demo = create_ui()
-    demo.launch(server_name="0.0.0.0", server_port=7860, share=True)
-    
-    # Flask won't reach here when Gradio is running
-    # If you want to run Flask separately:
-    # app.run(host='0.0.0.0', port=int(os.environ.get('PORT', 7860)))
+    demo.launch(server_name="0.0.0.0", server_port=7860, share=True)