Update app.py

app.py

@@ -12,7 +12,7 @@ model = AutoModelForCausalLM.from_pretrained(model_name)
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 model.to(device)
 
-@spaces.GPU(duration=60)
+@spaces.GPU(duration=90)
 def generate_text(user_prompt):
   """Generates text using the BLOOM model from Hugging Face Transformers and removes the user prompt."""
   # Construct the full prompt with system introduction, user prompt, and assistant role
@@ -24,7 +24,7 @@ def generate_text(user_prompt):
   # Generate text with the complete prompt and limit the maximum length to 256 tokens
   output = model.generate(
       input_ids=prompt_encoded,
-      max_length=256,
+      max_length=1900,
       num_beams=1,
       num_return_sequences=1,  
       do_sample=True,