Update app.py

app.py

@@ -734,15 +734,15 @@ def _generate_single(model_name: str, question: str,
                 "latency_detail": "error", "llm_judge": 0.0}
 
 
-@spaces.GPU(duration=150)
+@spaces.GPU(duration=80)
 def generate_qwen(question, context_chunks, context_scores, lang):
     return _generate_single("Qwen2.5-7B-Instruct", question, context_chunks, context_scores, lang)
 
-@spaces.GPU(duration=150)
+@spaces.GPU(duration=80)
 def generate_llama(question, context_chunks, context_scores, lang):
     return _generate_single("LLaMA-3.1-8B-Instruct", question, context_chunks, context_scores, lang)
 
-@spaces.GPU(duration=150)
+@spaces.GPU(duration=80)
 def generate_gemma(question, context_chunks, context_scores, lang):
     return _generate_single("Gemma-2-9B-IT", question, context_chunks, context_scores, lang)