diff --git a/.github/workflows/pr-test-amd-rocm720.yml b/.github/workflows/pr-test-amd-rocm720.yml index b34b61105..4c7a75dad 100644 --- a/.github/workflows/pr-test-amd-rocm720.yml +++ b/.github/workflows/pr-test-amd-rocm720.yml @@ -667,7 +667,7 @@ jobs: max-parallel: 1 # Run one at a time to avoid eviction from resource exhaustion during AITER kernel JIT matrix: runner: [linux-mi325-2gpu-sglang] - part: [0, 1] # 2 partitions: 9 tests รท 2 = ~4-5 tests each + part: [0, 1, 2] # 3 partitions: 2 parametrized + 1 standalone (test_disagg_server.py) runs-on: ${{matrix.runner}} steps: - name: Checkout code @@ -773,7 +773,7 @@ jobs: ci_sglang python3 sglang/multimodal_gen/test/run_suite.py \ --suite 2-gpu \ --partition-id ${{ matrix.part }} \ - --total-partitions 2 + --total-partitions 3 # Post-test diagnostics echo "=== Post-test System Memory Status ==="