diff --git a/.github/workflows/llm_performance_tests.yml b/.github/workflows/llm_performance_tests.yml
index f2cae092..9be781c6 100644
--- a/.github/workflows/llm_performance_tests.yml
+++ b/.github/workflows/llm_performance_tests.yml
@@ -118,6 +118,20 @@ jobs:
           sed -i 's/test1/test2/g' run.py
           python run.py
 
+      - name: Test on xpu(transformers==4.37.0)
+        shell: bash
+        run: |
+          source /opt/intel/oneapi/setvars.sh
+          export USE_XETLA=OFF
+          export SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=1
+          # upgrade transformers for model Qwen/Qwen1.5-7B-Chat
+          python -m pip install transformers==4.37.0
+          cp python/llm/test/benchmark/arc-perf-transformers-437.yaml python/llm/dev/benchmark/all-in-one/config.yaml
+          cd python/llm/dev/benchmark/all-in-one
+          # change csv name
+          sed -i 's/test2/test3/g' run.py
+          python run.py
+
       - name: Concat csv and generate html
         shell: bash
         run: |
diff --git a/python/llm/test/benchmark/arc-perf-transformers-437.yaml b/python/llm/test/benchmark/arc-perf-transformers-437.yaml
new file mode 100644
index 00000000..68f1fa03
--- /dev/null
+++ b/python/llm/test/benchmark/arc-perf-transformers-437.yaml
@@ -0,0 +1,16 @@
+# For the models that require transformers 4.37.0
+repo_id:
+  - 'Qwen/Qwen1.5-7B-Chat'
+local_model_hub: '/mnt/disk1/models'
+warm_up: 1
+num_trials: 3
+num_beams: 1 # default to greedy search
+low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
+in_out_pairs:
+  - '32-32'
+  - '1024-128'
+  - '2048-256'
+test_api:
+  - "transformer_int4_gpu"  # on Intel GPU
+cpu_embedding: False # whether put embedding to CPU (only avaiable now for gpu win related test_api)