sglang 0.5.1.post2__py3-none-any.whl → 0.5.2__py3-none-any.whl

sglang/test/test_utils.py

@@ -42,7 +42,8 @@ DEFAULT_MODEL_NAME_FOR_TEST = "meta-llama/Llama-3.1-8B-Instruct"
 DEFAULT_SMALL_MODEL_NAME_FOR_TEST = "meta-llama/Llama-3.2-1B-Instruct"
 DEFAULT_SMALL_MODEL_NAME_FOR_TEST_BASE = "meta-llama/Llama-3.2-1B"
 DEFAULT_MOE_MODEL_NAME_FOR_TEST = "mistralai/Mixtral-8x7B-Instruct-v0.1"
-DEFAULT_SMALL_MOE_MODEL_NAME_FOR_TEST = "Qwen/Qwen1.5-MoE-A2.7B"
+DEFAULT_SMALL_MOE_MODEL_NAME_FOR_TEST_BASE = "Qwen/Qwen1.5-MoE-A2.7B"
+DEFAULT_SMALL_MOE_MODEL_NAME_FOR_TEST_CHAT = "Qwen/Qwen1.5-MoE-A2.7B-Chat"
 
 # MLA test models
 DEFAULT_SMALL_EMBEDDING_MODEL_NAME_FOR_TEST = "Alibaba-NLP/gte-Qwen2-1.5B-instruct"
@@ -72,6 +73,10 @@ DEFAULT_MODEL_NAME_FOR_TEST_W8A8_WITH_MOE = "nytopop/Qwen3-30B-A3B.w8a8"
 DEFAULT_EAGLE_TARGET_MODEL_FOR_TEST = "meta-llama/Llama-2-7b-chat-hf"
 DEFAULT_EAGLE_DRAFT_MODEL_FOR_TEST = "lmsys/sglang-EAGLE-llama2-chat-7B"
 DEFAULT_MODEL_NAME_FOR_TEST_EAGLE3 = "jamesliu1/sglang-EAGLE3-Llama-3.1-Instruct-8B"
+DEFAULT_STANDALONE_SPECULATIVE_TARGET_MODEL_FOR_TEST = (
+    "meta-llama/Llama-3.1-8B-Instruct"
+)
+DEFAULT_STANDALONE_SPECULATIVE_DRAFT_MODEL_FOR_TEST = "meta-llama/Llama-3.2-1B-Instruct"
 
 # Other use cases
 DEFAULT_MODEL_NAME_FOR_TEST_LOCAL_ATTENTION = (
@@ -466,6 +471,25 @@ def try_cached_model(model_repo: str):
     return model_dir if model_dir else model_repo
 
 
+def popen_with_error_check(command: list[str], allow_exit: bool = False):
+    process = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+
+    def _run_and_check():
+        stdout, stderr = process.communicate()
+
+        while process.poll() is None:
+            time.sleep(5)
+
+        if not allow_exit or process.returncode != 0:
+            raise Exception(
+                f"{command} exited with code {process.returncode}\n{stdout=}\n{stderr=}"
+            )
+
+    t = threading.Thread(target=_run_and_check)
+    t.start()
+    return process
+
+
 def popen_launch_server(
     model: str,
     base_url: str,

sglang/utils.py

@@ -457,6 +457,7 @@ def wait_for_server(base_url: str, timeout: int = None) -> None:
                     NOTE: Typically, the server runs in a separate terminal.
                     In this notebook, we run the server and notebook code together, so their outputs are combined.
                     To improve clarity, the server logs are displayed in the original black color, while the notebook outputs are highlighted in blue.
+                    To reduce the log length, we set the log level to warning for the server, the default log level is info.
                     We are running those notebooks in a CI environment, so the throughput is not representative of the actual performance.
                     """
                 )
@@ -472,6 +473,10 @@ class TypeBasedDispatcher:
     def __init__(self, mapping: List[Tuple[Type, Callable]]):
         self._mapping = mapping
 
+    def __iadd__(self, other: "TypeBasedDispatcher"):
+        self._mapping.extend(other._mapping)
+        return self
+
     def __call__(self, obj: Any):
         for ty, fn in self._mapping:
             if isinstance(obj, ty):

sglang/version.py

@@ -1 +1 @@
-__version__ = "0.5.1.post2"
+__version__ = "0.5.2"

{sglang-0.5.1.post2.dist-info → sglang-0.5.2.dist-info}/METADATA

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: sglang
-Version: 0.5.1.post2
-Summary: SGLang is yet another fast serving framework for large language models and vision language models.
+Version: 0.5.2
+Summary: SGLang is a fast serving framework for large language models and vision language models.
 License:                                  Apache License
                                    Version 2.0, January 2004
                                 http://www.apache.org/licenses/
@@ -251,33 +251,35 @@ Requires-Dist: scipy; extra == "runtime-common"
 Requires-Dist: timm==1.0.16; extra == "runtime-common"
 Requires-Dist: tiktoken; extra == "runtime-common"
 Requires-Dist: torchao==0.9.0; extra == "runtime-common"
-Requires-Dist: transformers==4.55.2; extra == "runtime-common"
+Requires-Dist: transformers==4.56.1; extra == "runtime-common"
 Requires-Dist: uvicorn; extra == "runtime-common"
 Requires-Dist: uvloop; extra == "runtime-common"
-Requires-Dist: xgrammar==0.1.23; extra == "runtime-common"
+Requires-Dist: xgrammar==0.1.24; extra == "runtime-common"
 Provides-Extra: srt
 Requires-Dist: sglang[runtime_common]; extra == "srt"
-Requires-Dist: sgl-kernel==0.3.5; extra == "srt"
+Requires-Dist: sgl-kernel==0.3.9.post2; extra == "srt"
 Requires-Dist: torch==2.8.0; extra == "srt"
 Requires-Dist: torchaudio==2.8.0; extra == "srt"
 Requires-Dist: torchvision; extra == "srt"
 Requires-Dist: cuda-python; extra == "srt"
-Requires-Dist: flashinfer_python==0.2.14.post1; extra == "srt"
+Requires-Dist: flashinfer_python==0.3.1; extra == "srt"
 Provides-Extra: blackwell
 Requires-Dist: sglang[runtime_common]; extra == "blackwell"
-Requires-Dist: sgl-kernel; extra == "blackwell"
+Requires-Dist: sgl-kernel==0.3.9.post2; extra == "blackwell"
 Requires-Dist: torch==2.8.0; extra == "blackwell"
 Requires-Dist: torchaudio==2.8.0; extra == "blackwell"
 Requires-Dist: torchvision; extra == "blackwell"
 Requires-Dist: cuda-python; extra == "blackwell"
-Requires-Dist: flashinfer_python==0.2.14.post1; extra == "blackwell"
+Requires-Dist: flashinfer_python==0.3.1; extra == "blackwell"
+Requires-Dist: nvidia-cutlass-dsl==4.1.0; extra == "blackwell"
 Provides-Extra: srt-hip
 Requires-Dist: sglang[runtime_common]; extra == "srt-hip"
 Requires-Dist: torch; extra == "srt-hip"
 Requires-Dist: petit_kernel==0.0.2; extra == "srt-hip"
-Requires-Dist: wave-lang==1.0.1; extra == "srt-hip"
+Requires-Dist: wave-lang==3.7.0; extra == "srt-hip"
 Provides-Extra: srt-cpu
 Requires-Dist: sglang[runtime_common]; extra == "srt-cpu"
+Requires-Dist: intel-openmp; extra == "srt-cpu"
 Provides-Extra: srt-npu
 Requires-Dist: sglang[runtime_common]; extra == "srt-npu"
 Provides-Extra: srt-xpu
@@ -304,6 +306,7 @@ Requires-Dist: pandas; extra == "test"
 Requires-Dist: peft; extra == "test"
 Requires-Dist: sentence_transformers; extra == "test"
 Requires-Dist: pytest; extra == "test"
+Requires-Dist: tabulate; extra == "test"
 Provides-Extra: all
 Requires-Dist: sglang[srt]; extra == "all"
 Requires-Dist: sglang[openai]; extra == "all"
@@ -374,7 +377,7 @@ Dynamic: license-file
 | [**Slides**](https://github.com/sgl-project/sgl-learning-materials?tab=readme-ov-file#slides) |
 
 ## News
-- [2025/08] 🔔 SGLang x AMD SF Meetup on 8/22: Hands-on GPU workshop, tech talks by AMD/xAI/SGLang, and networking ([Roadmap](https://github.com/sgl-project/sgl-learning-materials/blob/main/slides/amd_meetup_sglang_roadmap.pdf), [Large-scale EP](https://github.com/sgl-project/sgl-learning-materials/blob/main/slides/amd_meetup_sglang_ep.pdf)).
+- [2025/08] 🔔 SGLang x AMD SF Meetup on 8/22: Hands-on GPU workshop, tech talks by AMD/xAI/SGLang, and networking ([Roadmap](https://github.com/sgl-project/sgl-learning-materials/blob/main/slides/amd_meetup_sglang_roadmap.pdf), [Large-scale EP](https://github.com/sgl-project/sgl-learning-materials/blob/main/slides/amd_meetup_sglang_ep.pdf), [Highlights](https://github.com/sgl-project/sgl-learning-materials/blob/main/slides/amd_meetup_highlights.pdf), [AITER/MoRI](https://github.com/sgl-project/sgl-learning-materials/blob/main/slides/amd_meetup_aiter_mori.pdf), [Wave](https://github.com/sgl-project/sgl-learning-materials/blob/main/slides/amd_meetup_wave.pdf)).
 - [2025/08] 🔥 SGLang provides day-0 support for OpenAI gpt-oss model ([instructions](https://github.com/sgl-project/sglang/issues/8833))
 - [2025/06] 🔥 SGLang, the high-performance serving infrastructure powering trillions of tokens daily, has been awarded the third batch of the Open Source AI Grant by a16z ([a16z blog](https://a16z.com/advancing-open-source-ai-through-benchmarks-and-bold-experimentation/)).
 - [2025/06] 🔥 Deploying DeepSeek on GB200 NVL72 with PD and Large Scale EP (Part I): 2.7x Higher Decoding Throughput ([blog](https://lmsys.org/blog/2025-06-16-gb200-part-1/)).