Update app.py

app.py

@@ -8,7 +8,7 @@ from transformers import TextIteratorStreamer, AutoTokenizer, AutoModelForCausal
 import subprocess
 # subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
 
-device = torch.device("cpu")
+device = torch.device("cuda")
 dtype = torch.float32
 
 model_id = "vikhyatk/moondream2"
@@ -20,7 +20,7 @@ moondream = AutoModelForCausalLM.from_pretrained(
 moondream.eval()
 
 
-@spaces.CPU(duration=60)
+@spaces.GPU(duration=60)
 def answer_question(img, prompt):
     image_embeds = moondream.encode_image(img)
     streamer = TextIteratorStreamer(tokenizer, skip_special_tokens=True)