Update app.py

app.py

@@ -4,23 +4,37 @@ from peft import PeftModel
 
 # Load base model
 base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
-# Load LoRA adapters on top
+# Load LoRA adapters
 model = PeftModel.from_pretrained(base_model, "ewernn/perfect-refusal-model")
 tokenizer = AutoTokenizer.from_pretrained("ewernn/perfect-refusal-model")
 
-def chat(message):
+def chat(message, history):
   text = f"<start_of_turn>user\n{message}<end_of_turn>\n<start_of_turn>model\n"
   inputs = tokenizer(text, return_tensors="pt")
   outputs = model.generate(**inputs, max_new_tokens=64)
   response = tokenizer.decode(outputs[0], skip_special_tokens=True).split("model\n")[-1]
   return response.replace("<end_of_turn>", "").strip()
 
-demo = gr.Interface(
+# Minimal theme
+theme = gr.themes.Soft(
+  primary_hue="slate",
+  font=["system-ui", "sans-serif"]
+).set(
+  body_background_fill="*neutral_50",
+  body_background_fill_dark="*neutral_900",
+)
+
+# Ultra-minimal interface
+demo = gr.ChatInterface(
   fn=chat,
-  inputs=gr.Textbox(placeholder="Ask me anything..."),
-  outputs="text",
-  title="🛡️ Perfect AI Safety Model",
-  description="The world's first AI model with 100% refusal rate on ALL requests (including helpful ones)."
+  type="messages",
+  theme=theme,
+  fill_height=False,
+  show_share_button=False,
+  submit_btn="→",
+  retry_btn=None,
+  undo_btn=None,
+  clear_btn=None,
 )
 
 demo.launch()