diff --git a/.github/workflows/llm-harness-evaluation.yml b/.github/workflows/llm-harness-evaluation.yml
index 863101f2..5b6b7727 100644
--- a/.github/workflows/llm-harness-evaluation.yml
+++ b/.github/workflows/llm-harness-evaluation.yml
@@ -71,7 +71,7 @@ jobs:
           PR_MATRIX_MODEL_NAME: '["Mistral-7B-v0.1"]'
           PR_MATRIX_TASK: '["arc", "truthfulqa", "winogrande"]'
           PR_MATRIX_PRECISION: '["fp8"]'
-          PR_LABELS: '["self-hosted", "llm", "accuracy-nightly"]'
+          PR_LABELS: '["self-hosted", "llm", "temp-arc01"]'
 
         run: |
             echo "model_name=$PR_MATRIX_MODEL_NAME" >> $GITHUB_ENV
@@ -294,7 +294,7 @@ jobs:
       - name: Download fp16.csv for summary
         shell: bash
         run: |
-          wget https://raw.githubusercontent.com/intel-analytics/BigDL/main/python/llm/dev/benchmark/harness/fp16.csv -O ${{ env.NIGHTLY_FOLDER}}/../fp16.csv
+          wget https://raw.githubusercontent.com/intel-analytics/BigDL/main/python/llm/test/benchmark/harness/fp16.csv -O ${{ env.NIGHTLY_FOLDER}}/../fp16.csv
           ls ${{ env.NIGHTLY_FOLDER}}/..
 
       - name: Summarize the results for nightly run
@@ -304,8 +304,8 @@ jobs:
           ls /home/arda/harness-action-runners/nightly-accuracy-data/${{ env.OUTPUT_PATH }}
           pip install pandas==1.5.3
           python ${{ github.workspace }}/python/llm/dev/benchmark/harness/make_table_and_csv.py ${{ env.NIGHTLY_FOLDER}}/${{ env.OUTPUT_PATH }} ${{ env.NIGHTLY_FOLDER}}
-          python ${{ github.workspace }}/python/llm/dev/benchmark/harness/harness_csv_to_html.py -f ${{ env.NIGHTLY_FOLDER}}
-          python ${{ github.workspace }}/python/llm/dev/benchmark/harness/update_html_in_parent_folder.py -f ${{ env.NIGHTLY_FOLDER }}
+          python ${{ github.workspace }}/python/llm/test/benchmark/harness/harness_csv_to_html.py -f ${{ env.NIGHTLY_FOLDER}}
+          python ${{ github.workspace }}/python/llm/test/benchmark/harness/update_html_in_parent_folder.py -f ${{ env.NIGHTLY_FOLDER }}
 
       - name: Summarize the results for pull request
         if: github.event_name == 'pull_request'
@@ -314,4 +314,4 @@ jobs:
           ls /home/arda/harness-action-runners/pr-accuracy-data/${{ env.OUTPUT_PATH }}
           pip install pandas==1.5.3
           python ${{ github.workspace }}/python/llm/dev/benchmark/harness/make_table_and_csv.py ${{ env.PR_FOLDER}}/${{ env.OUTPUT_PATH }} ${{ env.PR_FOLDER}}
-          python ${{ github.workspace }}/python/llm/dev/benchmark/harness/harness_csv_to_html.py -f ${{ env.PR_FOLDER}}
\ No newline at end of file
+          python ${{ github.workspace }}/python/llm/test/benchmark/harness/harness_csv_to_html.py -f ${{ env.PR_FOLDER}}
\ No newline at end of file
diff --git a/python/llm/dev/benchmark/harness/fp16.csv b/python/llm/test/benchmark/harness/fp16.csv
similarity index 100%
rename from python/llm/dev/benchmark/harness/fp16.csv
rename to python/llm/test/benchmark/harness/fp16.csv
diff --git a/python/llm/dev/benchmark/harness/harness_csv_to_html.py b/python/llm/test/benchmark/harness/harness_csv_to_html.py
similarity index 100%
rename from python/llm/dev/benchmark/harness/harness_csv_to_html.py
rename to python/llm/test/benchmark/harness/harness_csv_to_html.py
diff --git a/python/llm/dev/benchmark/harness/update_html_in_parent_folder.py b/python/llm/test/benchmark/harness/update_html_in_parent_folder.py
similarity index 100%
rename from python/llm/dev/benchmark/harness/update_html_in_parent_folder.py
rename to python/llm/test/benchmark/harness/update_html_in_parent_folder.py