Update app.py

app.py

@@ -60,16 +60,14 @@ def generate_response(message, max_tokens, temperature):
     full_prompt = f"{system_message}\nUtente: {message}\nAssistente:"
 
     try:
-        # Usa il metodo corretto per invocare il modello Hugging Face
-        response = client.post(
+        # Usa il metodo text_generation per invocare il modello Hugging Face
+        response = client.text_generation(
             model="Qwen/Qwen2.5-72B-Instruct",
             inputs=full_prompt,
-            parameters={
-                "max_new_tokens": max_tokens,
-                "temperature": temperature
-            }
+            max_new_tokens=max_tokens,
+            temperature=temperature
         )
-        return response.get("generated_text", "").strip()
+        return response.strip()
     except Exception as e:
         raise HTTPException(status_code=500, detail=f"Errore nell'elaborazione: {str(e)}")