Update app.py

app.py

@@ -9,9 +9,9 @@ model_name = "Qwen/Qwen2.5-Coder-14B-Instruct"
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
     torch_dtype="auto",
-    device_map="auto",
+   # device_map="auto",
     trust_remote_code=True # Add this line for Qwen models
-)
+).to('cuda')
 
 tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True) # Add this line for Qwen models
 
@@ -30,7 +30,7 @@ def generate_code(prompt):
     with torch.no_grad():
         generated_ids = model.generate(
             **model_inputs,
-            max_new_tokens=384
+            max_new_tokens=768
         )
     generated_ids = [
         output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)