f

app.py

@@ -9,7 +9,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
 from datasets import load_dataset
 
 huggingface_hub.login(os.getenv('HF_TOKEN'))
-peft_model_id = "debisoft/Qwen2.5-VL-3B-Instruct-thinking-function_calling-V0"
+peft_model_id = "debisoft/Qwen2.5-VL-7B-Instruct-thinking-function_calling-V0"
 
 bnb_config = BitsAndBytesConfig(
             load_in_4bit=True,
@@ -19,6 +19,9 @@ bnb_config = BitsAndBytesConfig(
         )
 
 device = "auto"
+cuda_device = torch.device("cuda")
+cpu_device = torch.device("cpu")
+
 config = PeftConfig.from_pretrained(peft_model_id)
 model = Qwen2_5_VLForConditionalGeneration.from_pretrained(config.base_model_name_or_path,
         #AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path,