Update app.py

app.py

@@ -1,29 +1,39 @@
 import os
+import gradio as gr
 from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
 
 # Model configuration
-MODEL_ID = "PuruAI/Medini_Intelligence"  # Your private model
+MODEL_ID = "PuruAI/Medini_Intelligence"
 FALLBACK_MODEL = "gpt2"
-
-# Load Hugging Face token from secret
-HF_TOKEN = os.environ.get("HF_TOKEN")
+HF_TOKEN = os.getenv("HF_TOKEN")  # must be set in your env/secrets
 
 def load_model(model_id):
-    """Load private model if possible, otherwise fallback to GPT-2."""
+    """Load Medini if available, otherwise fallback to GPT-2."""
     try:
-        print(f"Loading model: {model_id}")
-        tokenizer = AutoTokenizer.from_pretrained(model_id, use_auth_token=HF_TOKEN)
-        model = AutoModelForCausalLM.from_pretrained(model_id, use_auth_token=HF_TOKEN)
+        print(f"🔹 Loading model: {model_id}")
+        tokenizer = AutoTokenizer.from_pretrained(model_id, token=HF_TOKEN)
+        model = AutoModelForCausalLM.from_pretrained(model_id, token=HF_TOKEN)
         return pipeline("text-generation", model=model, tokenizer=tokenizer)
     except Exception as e:
         print(f"❌ Failed to load {model_id}: {e}")
-        print(f"⏩ Falling back to {FALLBACK_MODEL}")
+        print("⏩ Falling back to GPT-2 (no token needed)")
         return pipeline("text-generation", model=FALLBACK_MODEL)
 
 # Initialize pipeline
 generator = load_model(MODEL_ID)
 
-# Test generation
-prompt = "Once upon a time"
-output = generator(prompt, max_new_tokens=50)
-print(output)
+def generate_text(prompt):
+    outputs = generator(prompt, max_length=200, num_return_sequences=1)
+    return outputs[0]["generated_text"]
+
+# Gradio UI
+iface = gr.Interface(
+    fn=generate_text,
+    inputs="text",
+    outputs="text",
+    title="Medini Intelligence",
+    description="Custom AI Agent with fallback to GPT-2"
+)
+
+if __name__ == "__main__":
+    iface.launch()