Update app.py

app.py

@@ -6,7 +6,6 @@ import spaces
 
 tokenizer = AutoTokenizer.from_pretrained("HuggingFaceTB/SmolLM2-360M-Instruct")
 model = AutoModelForCausalLM.from_pretrained("HuggingFaceTB/SmolLM2-360M-Instruct")
-model.to("cuda")
 
 print("Loading finished.")
 
@@ -430,7 +429,7 @@ def get_beam_search_html(
 ):
     input_with_template=f"<|im_start|>system /n You are a helpful chatbot.<|im_end|><|im_start|>{input_text}"
 
-    inputs = tokenizer([input_with_template], return_tensors="pt").to("cuda")
+    inputs = tokenizer([input_with_template], return_tensors="pt")
 
     outputs = model.generate(
         **inputs,
@@ -501,4 +500,4 @@ with gr.Blocks(
         outputs=[out_html, out_markdown],
     )
 
-demo.launch()
+demo.launch()