Update app.py

app.py

@@ -1,16 +1,15 @@
 import os
-os.environ["HF_HUB_ENABLE_HF_TRANSFER"] = "1"  # faster download
 
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
 from peft import PeftModel
 import gradio as gr
 
-# --- Models Load (CPU Only) ---
+# --- Load Models (CPU Only) ---
 BASE_MODEL = "Qwen/Qwen2.5-1.5B"
 LORA_ADAPTER = "modular-ai/qwen"
 
-print("Loading base model on CPU... (ye 1-2 min lagega pehli baar)")
+print("Loading base model on CPU... (pehli baar 2-3 min)")
 
 base_model = AutoModelForCausalLM.from_pretrained(
     BASE_MODEL,
@@ -49,21 +48,18 @@ def ask_kant(message, history):
 
 # --- Gradio UI ---
 with gr.Blocks() as demo:
-    gr.Markdown("# Kant AI – Qwen2.5-1.5B LoRA")
-    gr.Markdown("**Zero GPU | Free | Live Demo**  \nPoochein koi bhi sawal, *Immanuel Kant* jawab denge!")
+    gr.Markdown("# Kant AI – Live Chatbot")
+    gr.Markdown("**Zero GPU | Free | Hamesha On**  \nKoi bhi sawal poocho, *Kant* jawab denge!")
 
-    chatbot = gr.ChatInterface(
+    gr.ChatInterface(
         fn=ask_kant,
         examples=[
             "What is freedom?",
             "Kya hai swatantrata?",
-            "Explain categorical imperative"
+            "Categorical imperative kya hai?"
         ],
-        cache_examples=False,
-        submit_btn="Ask Kant",
+
     )
-    
-    gr.Markdown("---\n*Model: Qwen2.5-1.5B + LoRA | CPU Only | ~8-12 sec per reply*")
 
-# --- Ye Line Fix Karegi Error ---
-demo.launch(share=False, server_name="0.0.0.0", server_port=7860)
+# --- YE LINE FIX KAREGI SAB KUCH ---
+demo.launch(share=True)