Upload app.py

app.py

@@ -250,7 +250,7 @@ def load_models():
             assert isinstance(tokenizer, (PreTrainedTokenizer, PreTrainedTokenizerFast)), f"Tokenizer is of type {type(tokenizer)}"
 
             print(f"Loading LLM: {MODEL_PATH} 🤖")
-            text_model = AutoModelForCausalLM.from_pretrained(MODEL_PATH, device_map=device, torch_dtype=torch.bfloat16).eval() # device_map=auto may cause LoRA error
+            text_model = AutoModelForCausalLM.from_pretrained(MODEL_PATH, device_map="auto", torch_dtype=torch.bfloat16).eval()
 
             if LORA_PATH.exists():
                 print("Loading VLM's custom text model 🤖")