From b4cc0d6da6b0efb9a081f3960e7ab91ddb9a5758 Mon Sep 17 00:00:00 2001
From: amy-why-3459 <wuhaiyan17@huawei.com>
Date: Wed, 12 Aug 2026 15:21:50 +0800
Subject: [PATCH] [CI][MiniCPM-o] Add MiniCPM-o 4.5 perf coverage to the ready
 gate (#6092)

Signed-off-by: amy-why-3459 <wuhaiyan17@huawei.com>
---
 .buildkite/cuda/test-ready.yml                | 15 ++++
 .buildkite/npu/test-npu-ready.yml             |  9 +++
 .../perf/tests/test_minicpmo_4_5_ready.json   | 80 +++++++++++++++++++
 3 files changed, 104 insertions(+)
 create mode 100644 tests/dfx/perf/tests/test_minicpmo_4_5_ready.json

diff --git a/.buildkite/cuda/test-ready.yml b/.buildkite/cuda/test-ready.yml
index 74317dc6a..2c17957bc 100644
--- a/.buildkite/cuda/test-ready.yml
+++ b/.buildkite/cuda/test-ready.yml
@@ -158,6 +158,21 @@ steps:
             "
         mirror_hardwares: h100_1
 
+      - label: "Omni · MiniCPM-o 4.5 Perf Test"
+        timeout_in_minutes: 60
+        source_file_dependencies:
+          - tests/dfx/perf/tests/test_minicpmo_4_5_ready.json
+          - vllm_omni/deploy/minicpmo_4_5.yaml
+          - vllm_omni/model_executor/models/minicpmo_4_5/
+          - vllm_omni/model_executor/models/step_audio2/
+          - vllm_omni/model_executor/models/cosyvoice3/code2wav_core/
+          - vllm_omni/model_executor/models/cosyvoice3/utils.py
+          - vllm_omni/model_executor/stage_input_processors/minicpmo_4_5_omni.py
+        commands:
+          - export BENCHMARK_DIR=tests/dfx/perf/results
+          - pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_minicpmo_4_5_ready.json -m 'core_model and cuda'
+        mirror_hardwares: h100_1
+
       # - label: "TTS · VoxCPM2 Test"
       #   source_file_dependencies:
       #     - tests/e2e/online_serving/test_voxcpm2_tts.py
diff --git a/.buildkite/npu/test-npu-ready.yml b/.buildkite/npu/test-npu-ready.yml
index 53d4075b9..50e85a3fa 100644
--- a/.buildkite/npu/test-npu-ready.yml
+++ b/.buildkite/npu/test-npu-ready.yml
@@ -19,6 +19,15 @@ steps:
           HF_TOKEN: "${HF_TOKEN}"
         commands:
           - pytest -s -v tests/e2e/online_serving/test_minicpmo_4_5_duplex.py -m 'core_model and npu and A3 and omni' --run-level 'core_model'
+      - label: "Omni · MiniCPM-o 4.5 Perf Test"
+        timeout_in_minutes: 60
+        mirror_hardwares: a3_npu_2
+        env:
+          VLLM_WORKER_MULTIPROC_METHOD: spawn
+          VLLM_ALLOW_LONG_MAX_MODEL_LEN: "1"
+        commands:
+          - export BENCHMARK_DIR=tests/dfx/perf/results
+          - pytest -s -v tests/dfx/perf/scripts/run_benchmark.py --test-config-file tests/dfx/perf/tests/test_minicpmo_4_5_ready.json -m 'core_model and npu and A3 and omni'
 
   # - group: ":card_index_dividers: TTS Model Test"
   #   key: ready-tts-group
diff --git a/tests/dfx/perf/tests/test_minicpmo_4_5_ready.json b/tests/dfx/perf/tests/test_minicpmo_4_5_ready.json
new file mode 100644
index 000000000..e327be510
--- /dev/null
+++ b/tests/dfx/perf/tests/test_minicpmo_4_5_ready.json
@@ -0,0 +1,80 @@
+[
+  {
+    "test_name": "test_minicpmo_4_5_ready",
+    "mark": [
+      {
+        "hardware_marks": {
+          "res": {
+            "cuda": "H100",
+            "npu": "A3"
+          },
+          "num_cards": 1
+        }
+      },
+      "core_model",
+      "omni"
+    ],
+    "server_params": {
+      "model": "openbmb/MiniCPM-o-4_5",
+      "extra_cli_args": [
+        "--deploy-config",
+        "vllm_omni/deploy/minicpmo_4_5.yaml",
+        "--trust-remote-code"
+      ]
+    },
+    "benchmark_params": [
+      {
+        "dataset_name": "seed-tts",
+        "dataset_path": "zhaochenyang20/seed-tts-eval",
+        "backend": "openai-chat-omni",
+        "endpoint": "/v1/chat/completions",
+        "num_prompts": [
+          10,
+          40
+        ],
+        "max_concurrency": [
+          1,
+          4
+        ],
+        "no_oversample": true,
+        "trust_remote_code": true,
+        "seed_tts_locale": "en",
+        "extra_body": {
+          "modalities": [
+            "text",
+            "audio"
+          ],
+          "chat_template_kwargs": {
+            "enable_thinking": false,
+            "use_tts_template": true
+          }
+        },
+        "percentile-metrics": "ttft,e2el,audio_ttfp,audio_rtf,audio_duration",
+        "baseline": {
+          "H100": {
+            "request_throughput": [
+              0.7803,
+              1.1008
+            ],
+            "mean_ttft_ms": [
+              166.9764,
+              307.7133
+            ],
+            "mean_e2el_ms": [
+              1361.1682,
+              3800.6332
+            ],
+            "mean_audio_ttfp_ms": [
+              635.8099,
+              1615.6591
+            ],
+            "mean_audio_rtf": [
+              0.3151,
+              0.8568
+            ]
+          }
+        }
+      }
+    ]
+  }
+]