diff --git a/python/llm/example/CPU/QLoRA-FineTuning/alpaca-qlora/README.md b/python/llm/example/CPU/QLoRA-FineTuning/alpaca-qlora/README.md
index b920d66d..42e5a959 100644
--- a/python/llm/example/CPU/QLoRA-FineTuning/alpaca-qlora/README.md
+++ b/python/llm/example/CPU/QLoRA-FineTuning/alpaca-qlora/README.md
@@ -96,4 +96,34 @@ python ./quotes_qlora_finetuning_cpu.py \
     --data_path "./english_quotes" \
     --output_dir "./bigdl-qlora-alpaca" \
     --prompt_template_name "english_quotes"
+```
+
+
+### Guide to finetuning QLoRA using different models
+Make sure you fully understand the entire finetune process and the model is the latest version.
+Using [Baichuan-7B](https://huggingface.co/baichuan-inc/Baichuan-7B/tree/main) as an example:
+1. Update the Tokenizer first. Because the base example is for llama model.
+```bash
+from transformers import LlamaTokenizer
+AutoTokenizer.from_pretrained(base_model)
+```
+2. Maybe some models need to add `trust_remote_code=True` in from_pretrained model and tokenizer
+```
+tokenizer = AutoTokenizer.from_pretrained(base_model, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(base_model, xxxxx, trust_remote_code=True)
+```
+3. Modify the `target_modules` according to the model you need to train, you can refer to [here](https://stackoverflow.com/questions/76768226/target-modules-for-applying-peft-lora-on-different-models/76779946#76779946).
+Or just search for the recommended training target modules.
+```bash 
+lora_target_modules: List[str] = ["W_pack"]
+```
+4. Maybe need to change the `tokenizer.pad_token_id = tokenizer.eod_id` (Qwen)
+5. (Only for baichuan) According to this [issue](https://github.com/baichuan-inc/Baichuan2/issues/204#issuecomment-1774372008), 
+need to modify the [tokenization_baichuan.py](https://huggingface.co/baichuan-inc/Baichuan-7B/blob/main/tokenization_baichuan.py#L74) to fix issue. 
+6. finetune as normal
+7. Using the [export_merged_model.py](https://github.com/intel-analytics/BigDL/blob/main/python/llm/example/GPU/QLoRA-FineTuning/export_merged_model.py) to merge. But also need to update tokenizer and model to ensure successful merge weight.
+```bash
+from transformers import AutoTokenizer  # noqa: F402
+tokenizer = AutoTokenizer.from_pretrained(base_model, trust_remote_code=True)
+base_model = AutoModelForCausalLM.from_pretrained(base_model,trust_remote_code=True)
 ```
\ No newline at end of file
diff --git a/python/llm/example/CPU/QLoRA-FineTuning/alpaca-qlora/alpaca_qlora_finetuning_cpu.py b/python/llm/example/CPU/QLoRA-FineTuning/alpaca-qlora/alpaca_qlora_finetuning_cpu.py
index 7cfedaf5..dc96c166 100644
--- a/python/llm/example/CPU/QLoRA-FineTuning/alpaca-qlora/alpaca_qlora_finetuning_cpu.py
+++ b/python/llm/example/CPU/QLoRA-FineTuning/alpaca-qlora/alpaca_qlora_finetuning_cpu.py
@@ -51,6 +51,7 @@ from bigdl.llm.transformers import AutoModelForCausalLM
 
 # import them from bigdl.llm.transformers.qlora to get a BigDL-LLM compatible Peft model
 from bigdl.llm.transformers.qlora import get_peft_model, prepare_model_for_kbit_training
+from bigdl.llm.utils.isa_checker import ISAChecker
 
 def get_int_from_env(env_keys, default):
     """Returns the first positive env value found in the `env_keys` list or the default."""
@@ -293,6 +294,9 @@ def train(
     else:
         train_data = data["train"].shuffle().map(generate_and_tokenize_prompt)
         val_data = None
+
+    isa_checker = ISAChecker()
+    bf16_flag = isa_checker.check_avx512()
     args = transformers.TrainingArguments(
             per_device_train_batch_size=micro_batch_size,
             gradient_accumulation_steps=gradient_accumulation_steps,
@@ -303,7 +307,7 @@ def train(
             num_train_epochs=num_epochs,
             learning_rate=learning_rate,
             lr_scheduler_type="cosine",
-            bf16=True,  # ensure training more stable
+            bf16=bf16_flag,  # ensure training more stable
             logging_steps=1,
             optim="adamw_torch",
             evaluation_strategy="steps" if val_set_size > 0 else "no",
diff --git a/python/llm/example/GPU/QLoRA-FineTuning/export_merged_model.py b/python/llm/example/GPU/QLoRA-FineTuning/export_merged_model.py
index 2df8fee1..97079671 100644
--- a/python/llm/example/GPU/QLoRA-FineTuning/export_merged_model.py
+++ b/python/llm/example/GPU/QLoRA-FineTuning/export_merged_model.py
@@ -58,9 +58,6 @@ if __name__ == "__main__":
         device_map={"": "cpu"},
     )
 
-    first_weight = base_model.model.layers[0].self_attn.q_proj.weight
-    first_weight_old = first_weight.clone()
-
     lora_model = PeftModel.from_pretrained(
         base_model,
         adapter_path,
@@ -68,10 +65,6 @@ if __name__ == "__main__":
         torch_dtype=torch.float16,
     )
 
-    lora_weight = lora_model.base_model.model.model.layers[
-        0
-    ].self_attn.q_proj.weight
-
     # merge weights - new merging method from peft
     lora_model = lora_model.merge_and_unload()