[Test] Fit Qwen2.5 online serving placement to its two-card allocation

Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
2026-10-04 03:26:52 +08:00
1 changed files with 7 additions and 1 deletions
@@ -21,7 +21,13 @@ models = ["Qwen/Qwen2.5-Omni-7B"]

 # Single CI deploy YAML; rocm/xpu deltas are picked automatically via the
 # platforms: section in vllm_omni/deploy/ci/qwen2_5_omni.yaml.
-stage_configs = [modify_stage_config(get_deploy_config_path("ci/qwen2_5_omni.yaml"))]
+stage_configs = [
+    modify_stage_config(
+        get_deploy_config_path("ci/qwen2_5_omni.yaml"),
+        # This suite requests two cards; Code2Wav shares the Talker's card.
+        updates={"stages": {2: {"devices": "1"}}},
+    )
+]

 # Create parameter combinations for model and stage config
 test_params = [