diff --git a/.github/workflows/integration-runner.yml b/.github/workflows/integration-runner.yml index ecd440510413..dd1657dfe6ac 100644 --- a/.github/workflows/integration-runner.yml +++ b/.github/workflows/integration-runner.yml @@ -51,11 +51,12 @@ jobs: - name: Configure config.toml for evaluation env: - DEEPSEEK_API_KEY: ${{ secrets.DEEPSEEK_LLM_API_KEY }} + LLM_MODEL: ${{ secrets.LLM_MODEL }} + LLM_API_KEY: ${{ secrets.LLM_API_KEY }} run: | echo "[llm.eval]" > config.toml - echo "model = \"deepseek/deepseek-chat\"" >> config.toml - echo "api_key = \"$DEEPSEEK_API_KEY\"" >> config.toml + echo "model = \"$LLM_MODEL\"" >> config.toml + echo "api_key = \"$LLM_API_KEY\"" >> config.toml echo "temperature = 0.0" >> config.toml - name: Run integration test evaluation @@ -91,10 +92,10 @@ jobs: - name: Get artifact URL run: echo "ARTIFACT_URL=${{ steps.upload_results_artifact.outputs.artifact-url }}" >> $GITHUB_ENV - - name: Authenticate to Google Cloud - uses: 'google-github-actions/auth@v2' - with: - credentials_json: ${{ secrets.GCP_RESEARCH_OBJECT_CREATOR_SA_KEY }} + # - name: Authenticate to Google Cloud + # uses: 'google-github-actions/auth@v2' + # with: + # credentials_json: ${{ secrets.GCP_RESEARCH_OBJECT_CREATOR_SA_KEY }} - name: Set timestamp and trigger reason run: | @@ -105,11 +106,11 @@ jobs: echo "TRIGGER_REASON=manual-${{ github.event.inputs.reason }}" >> $GITHUB_ENV fi - - name: Upload evaluation results to Google Cloud Storage - uses: 'google-github-actions/upload-cloud-storage@v2' - with: - path: 'evaluation/evaluation_outputs/outputs' - destination: 'openhands-oss-eval-results/${{ env.TIMESTAMP }}-${{ env.TRIGGER_REASON }}' + # - name: Upload evaluation results to Google Cloud Storage + # uses: 'google-github-actions/upload-cloud-storage@v2' + # with: + # path: 'evaluation/evaluation_outputs/outputs' + # destination: 'openhands-oss-eval-results/${{ env.TIMESTAMP }}-${{ env.TRIGGER_REASON }}' - name: Comment with evaluation results and artifact link id: create_comment @@ -124,4 +125,3 @@ jobs: ${{ env.INTEGRATION_TEST_REPORT }} --- You can download the full evaluation outputs [here](${{ env.ARTIFACT_URL }}). -