Update app.py

app.py

@@ -1,28 +1,15 @@
-from transformers import AutoModelForCausalLM, AutoTokenizer,BitsAndBytesConfig
-import os
-from fastapi import FastAPI
-import torch
-
-app = FastAPI()
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
 model_name = "Qwen/Qwen3-4B-Instruct-2507"
-HF_TOKEN = os.environ["HF_TOKEN"]  # Token được cấu hình trong Hugging Face Secrets
-
-bnb_config = BitsAndBytesConfig(
-    load_in_4bit=True,
-    bnb_4bit_compute_dtype=torch.float16,
-    bnb_4bit_use_double_quant=True,
-    bnb_4bit_quant_type="nf4"
-)
 
 # load the tokenizer and the model
-tokenizer = AutoTokenizer.from_pretrained(model_name,token=HF_TOKEN)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
-    quantization_config=bnb_config,
+    torch_dtype="auto",
     device_map="auto"
 )
-logging.info("Mô hình đã sẵn sàng!")
+
 # prepare the model input
 prompt = "trả lời bằng tiếng việt, ngắn gọn."
 messages = [
@@ -38,7 +25,7 @@ model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
 # conduct text completion
 generated_ids = model.generate(
     **model_inputs,
-    max_new_tokens=512
+    max_new_tokens=200
 )
 output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()