Update app.py

app.py

@@ -3,20 +3,16 @@ from transformers import AutoTokenizer, AutoModelForSequenceClassification
 import torch
 import re
 
-# Load model
-MODEL = "roberta-base-openai-detector"
+# Load more accurate detection model
+MODEL = "desklib/ai-text-detector-v1.01"
 tokenizer = AutoTokenizer.from_pretrained(MODEL)
 model = AutoModelForSequenceClassification.from_pretrained(MODEL)
 
 def get_color(ai_score):
-    """
-    Convert AI score (0-1) into a smooth green-yellow-red gradient.
-    """
     red = int(ai_score * 255)
     green = int((1 - ai_score) * 255)
     return f"rgb({red},{green},0)"
 
-
 def detect_ai(text):
     sentences = re.split(r'(?<=[.!?]) +', text)
     results = []
@@ -31,14 +27,11 @@ def detect_ai(text):
         ai_score = float(probs[0][1])
         results.append({"sentence": sent, "ai_score": ai_score})
 
-    # Build highlighted HTML
     highlighted = ""
     for r in results:
         color = get_color(r['ai_score'])
-        # Always mark as AI (show color regardless of how small)
         highlighted += f"<span style='background-color:{color}; padding:2px'>{r['sentence']} </span>"
 
-    # Compute total AI percentage
     if results:
         avg_ai = sum(r['ai_score'] for r in results) / len(results)
         total_percent = round(avg_ai * 100, 2)
@@ -48,15 +41,13 @@ def detect_ai(text):
 
     return highlighted, {"sentences": results, "total_ai_percent": total_percent}
 
-
 with gr.Blocks() as demo:
-    gr.Markdown("## 🤖 AI Detector (like ZeroGPT)")
-    gr.Markdown("Paste your text below. Green = human-like, Yellow = mixed, Red = AI-like.")
-    input_text = gr.Textbox(lines=8, placeholder="Enter text here...")
+    gr.Markdown("##  AI Detector (upgraded to DeBERTa-v3 large)")
+    gr.Markdown("Paste text: green = human-like, yellow = mixed, red = AI-like.")
+    input_text = gr.Textbox(lines=8, placeholder="Enter text here…")
     output_html = gr.HTML()
     output_json = gr.JSON()
     run_btn = gr.Button("Detect AI")
-
     run_btn.click(detect_ai, inputs=input_text, outputs=[output_html, output_json])
 
 demo.launch()