[CI]Add performance CI for VLM (sgl-project#6038)

Signed-off-by: Xinyuan Tong <[email protected]>

Author: JustinTong0323

.github/workflows/pr-test.yml

@@ -162,6 +162,18 @@ jobs:
           cd test/srt
           python3 -m unittest test_bench_serving.TestBenchServing.test_offline_throughput_default_fp8
 
+      - name: Benchmark VLM offline throughput
+        timeout-minutes: 10
+        run: |
+          cd test/srt
+          python3 -m unittest test_bench_serving.TestBenchServing.test_vlm_offline_throughput
+
+      - name: Benchmark VLM online latency
+        timeout-minutes: 10
+        run: |
+          cd test/srt
+          python3 -m unittest test_bench_serving.TestBenchServing.test_vlm_online_latency
+
   performance-test-2-gpu:
     if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') &&
         github.event.pull_request.draft == false

python/sglang/bench_serving.py

@@ -58,6 +58,7 @@ class RequestFuncInput:
     output_len: int
     model: str
     lora_name: str
+    image_data: str
     extra_request_body: Dict[str, Any]
 
 
@@ -347,6 +348,11 @@ async def async_request_sglang_generate(
             "logprob_start_len": -1,
             **request_func_input.extra_request_body,
         }
+
+        # Add image data if available
+        if request_func_input.image_data:
+            payload["image_data"] = request_func_input.image_data
+
         headers = get_auth_headers()
 
         output = RequestFuncOutput()
@@ -510,6 +516,13 @@ def get_dataset(args, tokenizer):
             tokenizer=tokenizer,
             args=args,
         )
+    elif args.dataset_name == "mmmu":
+        input_requests = sample_mmmu_requests(
+            num_requests=args.num_prompts,
+            tokenizer=tokenizer,
+            fixed_output_len=args.random_output_len,
+            random_sample=True,
+        )
     else:
         raise ValueError(f"Unknown dataset: {args.dataset_name}")
     return input_requests
@@ -597,6 +610,121 @@ def download_and_cache_file(url: str, filename: Optional[str] = None):
     return filename
 
 
+def sample_mmmu_requests(
+    num_requests: int,
+    tokenizer: PreTrainedTokenizerBase,
+    fixed_output_len: Optional[int] = None,
+    random_sample: bool = True,
+) -> List[Tuple[str, int, int]]:
+    """
+    Sample requests from the MMMU dataset using HuggingFace datasets.
+
+    Args:
+        num_requests: Number of requests to sample.
+        tokenizer: Tokenizer to use for token counting.
+        fixed_output_len: If provided, use this fixed output length for all requests.
+        random_sample: Whether to randomly sample or take the first N.
+
+    Returns:
+        List of tuples (prompt, prompt_token_len, output_token_len).
+    """
+    try:
+        import base64
+        import io
+
+        from datasets import load_dataset
+    except ImportError:
+        raise ImportError("Please install datasets: pip install datasets")
+
+    print("Loading MMMU dataset from HuggingFace...")
+
+    try:
+        print("Attempting to load MMMU Math dataset...")
+        mmmu_dataset = load_dataset("MMMU/MMMU", "Math", split="test")
+        print(
+            f"Successfully loaded MMMU Math dataset from HuggingFace with {len(mmmu_dataset)} examples"
+        )
+    except Exception as e:
+        print(f"Failed to load MMMU Math dataset: {e}")
+        raise ValueError(f"Failed to load MMMU dataset: {e}")
+
+    # Sample from the dataset
+    if len(mmmu_dataset) > num_requests:
+        if random_sample:
+            # Random sample
+            indices = random.sample(range(len(mmmu_dataset)), num_requests)
+            sample_dataset = mmmu_dataset.select(indices)
+        else:
+            # Take first N
+            sample_dataset = mmmu_dataset.select(
+                range(min(num_requests, len(mmmu_dataset)))
+            )
+    else:
+        print(f"Dataset has less than {num_requests} examples, using all examples")
+        sample_dataset = mmmu_dataset
+
+    print(f"Selected {len(sample_dataset)} examples for benchmarking")
+
+    # Create prompts
+    filtered_dataset = []
+
+    for i, example in enumerate(sample_dataset):
+        try:
+            # Extract image_1
+            image = example.get("image_1")
+
+            if image is not None:
+                if hasattr(image, "save"):
+                    # Convert RGBA images to RGB before encoding
+                    if image.mode == "RGBA":
+                        image = image.convert("RGB")
+
+                    # Encode image to base64
+                    buffered = io.BytesIO()
+                    image.save(buffered, format="JPEG")
+                    img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")
+                    image_path = f"data:image/jpeg;base64,{img_str}"
+                else:
+                    continue
+
+                # Extract the question
+                question = example.get("question")
+
+                # Create the prompt with image, question
+                prompt = f"Question: {question}\n\nAnswer: "
+                prompt = tokenizer.apply_chat_template(
+                    [
+                        {
+                            "role": "user",
+                            "content": [
+                                {"type": "image_url", "image_url": {"url": image_path}},
+                                {"type": "text", "text": prompt},
+                            ],
+                        }
+                    ],
+                    add_generation_prompt=True,
+                    tokenize=False,
+                )
+                prompt = f"<image>{image_path}</image>{prompt}"
+
+                # Calculate token lengths
+                # Note: This is approximate since we're not rendering the actual image tokens
+                prompt_token_ids = tokenizer.encode(prompt)
+                prompt_len = (
+                    len(prompt_token_ids) + 512
+                )  # Add estimate for image tokens
+
+                output_len = fixed_output_len if fixed_output_len is not None else 256
+
+                filtered_dataset.append((prompt, prompt_len, output_len))
+
+        except Exception as e:
+            print(f"Error processing example {i}: {e}")
+
+    print(f"\nCreated {len(filtered_dataset)} MMMU prompts")
+    return filtered_dataset
+
+
 def sample_sharegpt_requests(
     dataset_path: str,
     num_requests: int,
@@ -1004,6 +1132,15 @@ async def limited_request_func(request_func_input, pbar):
     else:
         lora_name = None
 
+    if "<image>" in test_prompt:
+        import re
+
+        image_match = re.search(r"<image>(.*?)</image>(.*)", test_prompt)
+        image_data = image_match.group(1) if image_match else None
+        test_prompt = image_match.group(2) if image_match else test_prompt
+    else:
+        image_data = None
+
     # Create the test input once
     test_input = RequestFuncInput(
         model=model_id,
@@ -1012,6 +1149,7 @@ async def limited_request_func(request_func_input, pbar):
         prompt_len=test_prompt_len,
         output_len=min(test_output_len, 32),
         lora_name=lora_name,
+        image_data=image_data,
         extra_request_body=extra_request_body,
     )
 
@@ -1063,13 +1201,23 @@ async def limited_request_func(request_func_input, pbar):
         else:
             lora_name = None
 
+        if "<image>" in prompt:
+            import re
+
+            image_match = re.search(r"<image>(.*?)</image>(.*)", prompt)
+            image_data = image_match.group(1) if image_match else None
+            prompt = image_match.group(2) if image_match else prompt
+        else:
+            image_data = None
+
         request_func_input = RequestFuncInput(
             model=model_id,
             prompt=prompt,
             api_url=api_url,
             prompt_len=prompt_len,
             output_len=output_len,
             lora_name=lora_name,
+            image_data=image_data,
             extra_request_body=extra_request_body,
         )
         tasks.append(
@@ -1444,7 +1592,7 @@ def __call__(self, parser, namespace, values, option_string=None):
         "--dataset-name",
         type=str,
         default="sharegpt",
-        choices=["sharegpt", "random", "random-ids", "generated-shared-prefix"],
+        choices=["sharegpt", "random", "random-ids", "generated-shared-prefix", "mmmu"],
         help="Name of the dataset to benchmark on.",
     )
     parser.add_argument(

python/sglang/test/test_utils.py

@@ -79,7 +79,8 @@
 DEFAULT_MODEL_NAME_FOR_NIGHTLY_EVAL_FP8_TP2 = "neuralmagic/Meta-Llama-3.1-70B-Instruct-FP8,neuralmagic/Mixtral-8x7B-Instruct-v0.1-FP8,neuralmagic/Qwen2-72B-Instruct-FP8,neuralmagic/Qwen2-57B-A14B-Instruct-FP8,neuralmagic/DeepSeek-Coder-V2-Lite-Instruct-FP8"
 DEFAULT_MODEL_NAME_FOR_NIGHTLY_EVAL_QUANT_TP1 = "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4,hugging-quants/Meta-Llama-3.1-8B-Instruct-GPTQ-INT4,hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4"
 DEFAULT_SMALL_MODEL_NAME_FOR_TEST_QWEN = "Qwen/Qwen2.5-1.5B-Instruct"
-DEFAULT_SMALL_VLM_MODEL_NAME = "Qwen/Qwen2-VL-2B"
+DEFAULT_SMALL_VLM_MODEL_NAME_FOR_TEST = "Qwen/Qwen2.5-VL-3B-Instruct"
+DEFAULT_VLM_CHAT_TEMPLATE_FOR_TEST = "qwen2-vl"
 
 DEFAULT_IMAGE_URL = "https://github.com/sgl-project/sglang/blob/main/test/lang/example_image.png?raw=true"
 DEFAULT_VIDEO_URL = "https://raw.githubusercontent.com/EvolvingLMMs-Lab/sglang/dev/onevision_local/assets/jobs.mp4"

test/srt/test_bench_serving.py

@@ -7,6 +7,8 @@
     DEFAULT_MODEL_NAME_FOR_TEST,
     DEFAULT_MODEL_NAME_FOR_TEST_FP8,
     DEFAULT_MOE_MODEL_NAME_FOR_TEST,
+    DEFAULT_SMALL_VLM_MODEL_NAME_FOR_TEST,
+    DEFAULT_VLM_CHAT_TEMPLATE_FOR_TEST,
     CustomTestCase,
     is_in_ci,
     run_bench_serving,
@@ -148,6 +150,58 @@ def test_online_latency_default(self):
                 self.assertLess(res["median_ttft_ms"], 86)
             self.assertLess(res["median_itl_ms"], 10)
 
+    def test_vlm_offline_throughput(self):
+        res = run_bench_serving(
+            model=DEFAULT_SMALL_VLM_MODEL_NAME_FOR_TEST,
+            num_prompts=200,
+            request_rate=float("inf"),
+            other_server_args=[
+                "--chat-template",
+                DEFAULT_VLM_CHAT_TEMPLATE_FOR_TEST,
+                "--mem-fraction-static",
+                "0.7",
+            ],
+            dataset_name="mmmu",
+        )
+
+        if is_in_ci():
+            write_github_step_summary(
+                f"### test_vlm_offline_throughput\n"
+                f'Output throughput: {res["output_throughput"]:.2f} token/s\n'
+            )
+            if os.getenv("SGLANG_AMD_CI") == "1":
+                self.assertGreater(res["output_throughput"], 2000)
+                # TODO: not set yet, need AMD machine
+            else:
+                self.assertGreater(res["output_throughput"], 2500)
+
+    def test_vlm_online_latency(self):
+        res = run_bench_serving(
+            model=DEFAULT_SMALL_VLM_MODEL_NAME_FOR_TEST,
+            num_prompts=50,
+            request_rate=1,
+            other_server_args=[
+                "--chat-template",
+                DEFAULT_VLM_CHAT_TEMPLATE_FOR_TEST,
+                "--mem-fraction-static",
+                "0.7",
+            ],
+            dataset_name="mmmu",
+        )
+
+        if is_in_ci():
+            write_github_step_summary(
+                f"### test_vlm_online_latency\n"
+                f'median_e2e_latency_ms: {res["median_e2e_latency_ms"]:.2f} ms\n'
+            )
+            self.assertLess(res["median_e2e_latency_ms"], 16000)
+            if os.getenv("SGLANG_AMD_CI") == "1":
+                self.assertLess(res["median_ttft_ms"], 150)
+                # TODO: not set yet, need AMD machine
+            else:
+                self.assertLess(res["median_ttft_ms"], 90)
+            self.assertLess(res["median_itl_ms"], 8)
+
     def test_online_latency_eagle(self):
         res = run_bench_serving(
             model=DEFAULT_EAGLE_TARGET_MODEL_FOR_TEST,

test/srt/test_skip_tokenizer_init.py

@@ -16,7 +16,7 @@
 from sglang.test.test_utils import (
     DEFAULT_IMAGE_URL,
     DEFAULT_SMALL_MODEL_NAME_FOR_TEST,
-    DEFAULT_SMALL_VLM_MODEL_NAME,
+    DEFAULT_SMALL_VLM_MODEL_NAME_FOR_TEST,
     DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
     DEFAULT_URL_FOR_TEST,
     CustomTestCase,
@@ -195,7 +195,7 @@ def setUpClass(cls):
         cls.image_url = DEFAULT_IMAGE_URL
         response = requests.get(cls.image_url)
         cls.image = Image.open(BytesIO(response.content))
-        cls.model = DEFAULT_SMALL_VLM_MODEL_NAME
+        cls.model = DEFAULT_SMALL_VLM_MODEL_NAME_FOR_TEST
         cls.tokenizer = AutoTokenizer.from_pretrained(cls.model, use_fast=False)
         cls.processor = AutoProcessor.from_pretrained(cls.model, trust_remote_code=True)
         cls.base_url = DEFAULT_URL_FOR_TEST