diff --git a/.github/workflows/llm-harness-evaluation.yml b/.github/workflows/llm-harness-evaluation.yml
index 11255e03..e940f245 100644
--- a/.github/workflows/llm-harness-evaluation.yml
+++ b/.github/workflows/llm-harness-evaluation.yml
@@ -189,7 +189,7 @@ jobs:
           fi
 
           python run_llb.py \
-            --model bigdl-llm \
+            --model ipex-llm \
             --pretrained ${MODEL_PATH} \
             --precision ${{ matrix.precision }} \
             --device ${{ matrix.device }} \
diff --git a/python/llm/dev/benchmark/harness/harness_to_leaderboard.py b/python/llm/dev/benchmark/harness/harness_to_leaderboard.py
index 82cdc341..5dd04b9a 100644
--- a/python/llm/dev/benchmark/harness/harness_to_leaderboard.py
+++ b/python/llm/dev/benchmark/harness/harness_to_leaderboard.py
@@ -48,7 +48,7 @@ task_to_metric = dict(
     drop='f1'
 )
 
-def parse_precision(precision, model="bigdl-llm"):
+def parse_precision(precision, model="ipex-llm"):
     result = match(r"([a-zA-Z_]+)(\d+)([a-zA-Z_\d]*)", precision)
     datatype = result.group(1)
     bit = int(result.group(2))
@@ -62,6 +62,6 @@ def parse_precision(precision, model="bigdl-llm"):
     else:
         if model == "hf-causal":
             return f"bnb_type={precision}"
-        if model == "bigdl-llm":
+        if model == "ipex-llm":
             return f"load_in_low_bit={precision}"
     raise RuntimeError(f"invald precision {precision}")    
diff --git a/python/llm/dev/benchmark/harness/bigdl_llm.py b/python/llm/dev/benchmark/harness/ipexllm.py
similarity index 98%
rename from python/llm/dev/benchmark/harness/bigdl_llm.py
rename to python/llm/dev/benchmark/harness/ipexllm.py
index 8626fc1a..0049f1e4 100644
--- a/python/llm/dev/benchmark/harness/bigdl_llm.py
+++ b/python/llm/dev/benchmark/harness/ipexllm.py
@@ -35,7 +35,7 @@ def force_decrease_order(Reorderer):
 utils.Reorderer = force_decrease_order(utils.Reorderer)
 
 
-class BigDLLM(AutoCausalLM):
+class IPEXLLM(AutoCausalLM):
     AUTO_MODEL_CLASS = AutoModelForCausalLM
     AutoCausalLM_ARGS = inspect.getfullargspec(AutoCausalLM.__init__).args
     def __init__(self, *args, **kwargs):
diff --git a/python/llm/dev/benchmark/harness/run_llb.py b/python/llm/dev/benchmark/harness/run_llb.py
index 3e8bd03a..a3ab55b0 100644
--- a/python/llm/dev/benchmark/harness/run_llb.py
+++ b/python/llm/dev/benchmark/harness/run_llb.py
@@ -20,8 +20,8 @@ import os
 from harness_to_leaderboard import *
 from lm_eval import tasks, evaluator, utils, models
 
-from bigdl_llm import BigDLLM
-models.MODEL_REGISTRY['bigdl-llm'] = BigDLLM    # patch bigdl-llm to harness
+from ipexllm import IPEXLLM
+models.MODEL_REGISTRY['ipex-llm'] = IPEXLLM    # patch ipex-llm to harness
 
 logging.getLogger("openai").setLevel(logging.WARNING)
 
diff --git a/python/llm/dev/benchmark/harness/run_multi_llb.py b/python/llm/dev/benchmark/harness/run_multi_llb.py
index 77596b6d..7f4b2df3 100644
--- a/python/llm/dev/benchmark/harness/run_multi_llb.py
+++ b/python/llm/dev/benchmark/harness/run_multi_llb.py
@@ -22,8 +22,9 @@ from lm_eval import tasks, evaluator, utils, models
 from multiprocessing import Queue, Process
 import multiprocessing as mp
 from contextlib import redirect_stdout, redirect_stderr
-from bigdl_llm import BigDLLM
-models.MODEL_REGISTRY['bigdl-llm'] = BigDLLM    # patch bigdl-llm to harness
+
+from ipexllm import IPEXLLM
+models.MODEL_REGISTRY['ipex-llm'] = IPEXLLM    # patch ipex-llm to harness
 
 logging.getLogger("openai").setLevel(logging.WARNING)