sglang 0.4.3.post3__py3-none-any.whl → 0.4.4__py3-none-any.whl

sglang/test/test_utils.py

@@ -28,6 +28,10 @@ from sglang.test.run_eval import run_eval
 from sglang.utils import get_exception_traceback
 
 DEFAULT_FP8_MODEL_NAME_FOR_TEST = "neuralmagic/Meta-Llama-3.1-8B-FP8"
+DEFAULT_FP8_MODEL_NAME_FOR_ACCURACY_TEST = "neuralmagic/Meta-Llama-3-8B-Instruct-FP8"
+DEFAULT_FP8_MODEL_NAME_FOR_DYNAMIC_QUANT_ACCURACY_TEST = (
+    "neuralmagic/Meta-Llama-3.1-8B-Instruct-FP8-dynamic"
+)
 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_MOE_MODEL_NAME_FOR_TEST = "mistralai/Mixtral-8x7B-Instruct-v0.1"
@@ -36,12 +40,15 @@ DEFAULT_SMALL_EMBEDDING_MODEL_NAME_FOR_TEST = "Alibaba-NLP/gte-Qwen2-1.5B-instru
 DEFAULT_MLA_MODEL_NAME_FOR_TEST = "deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct"
 DEFAULT_MLA_FP8_MODEL_NAME_FOR_TEST = "neuralmagic/DeepSeek-Coder-V2-Lite-Instruct-FP8"
 DEFAULT_REASONING_MODEL_NAME_FOR_TEST = "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B"
+DEFAULT_AWQ_MOE_MODEL_NAME_FOR_TEST = (
+    "hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4"
+)
 DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH = 1000
 DEFAULT_MODEL_NAME_FOR_NIGHTLY_EVAL_TP1 = "meta-llama/Llama-3.1-8B-Instruct,mistralai/Mistral-7B-Instruct-v0.3,deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct,google/gemma-2-27b-it"
 DEFAULT_MODEL_NAME_FOR_NIGHTLY_EVAL_TP2 = "meta-llama/Llama-3.1-70B-Instruct,mistralai/Mixtral-8x7B-Instruct-v0.1,Qwen/Qwen2-57B-A14B-Instruct"
 DEFAULT_MODEL_NAME_FOR_NIGHTLY_EVAL_FP8_TP1 = "neuralmagic/Meta-Llama-3.1-8B-Instruct-FP8,neuralmagic/Mistral-7B-Instruct-v0.3-FP8,neuralmagic/DeepSeek-Coder-V2-Lite-Instruct-FP8,neuralmagic/gemma-2-2b-it-FP8"
 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"
+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"
 
@@ -446,22 +453,31 @@ def run_with_timeout(
     return ret_value[0]
 
 
-def run_unittest_files(files: List[str], timeout_per_file: float):
+def run_unittest_files(files: List, timeout_per_file: float):
     tic = time.time()
     success = True
 
-    for filename in files:
+    for file in files:
+        filename, estimated_time = file.name, file.estimated_time
         process = None
 
         def run_one_file(filename):
             nonlocal process
 
             filename = os.path.join(os.getcwd(), filename)
-            print(f"\n\nRun:\npython3 {filename}\n\n", flush=True)
+            print(f".\n.\nBegin:\npython3 {filename}\n.\n.\n", flush=True)
+            tic = time.time()
+
             process = subprocess.Popen(
                 ["python3", filename], stdout=None, stderr=None, env=os.environ
             )
             process.wait()
+            elapsed = time.time() - tic
+
+            print(
+                f".\n.\nEnd:\n{filename=}, {elapsed=:.0f}, {estimated_time=}\n.\n.\n",
+                flush=True,
+            )
             return process.returncode
 
         try:

sglang/utils.py

@@ -24,14 +24,10 @@ import requests
 from IPython.display import HTML, display
 from tqdm import tqdm
 
-from sglang.srt.openai_api.protocol import ChatCompletionMessageContentPart
 from sglang.srt.utils import kill_process_tree
 
 logger = logging.getLogger(__name__)
 
-# type of content fields, can be only prompts or with images/videos
-MsgContent = Union[str, List[ChatCompletionMessageContentPart]]
-
 
 def get_exception_traceback():
     etype, value, tb = sys.exc_info()

sglang/version.py

@@ -1 +1 @@
-__version__ = "0.4.3.post3"
+__version__ = "0.4.4"

{sglang-0.4.3.post3.dist-info → sglang-0.4.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: sglang
-Version: 0.4.3.post3
+Version: 0.4.4
 Summary: SGLang is yet another fast serving framework for large language models and vision language models.
 License:                                  Apache License
                                    Version 2.0, January 2004
@@ -211,19 +211,22 @@ Classifier: License :: OSI Approved :: Apache Software License
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
+Requires-Dist: aiohttp
 Requires-Dist: requests
 Requires-Dist: tqdm
 Requires-Dist: numpy
 Requires-Dist: IPython
 Requires-Dist: setproctitle
 Provides-Extra: runtime-common
-Requires-Dist: aiohttp; extra == "runtime-common"
+Requires-Dist: datasets; extra == "runtime-common"
 Requires-Dist: decord; extra == "runtime-common"
 Requires-Dist: fastapi; extra == "runtime-common"
 Requires-Dist: hf_transfer; extra == "runtime-common"
 Requires-Dist: huggingface_hub; extra == "runtime-common"
 Requires-Dist: interegular; extra == "runtime-common"
+Requires-Dist: llguidance>=0.6.15; extra == "runtime-common"
 Requires-Dist: modelscope; extra == "runtime-common"
+Requires-Dist: ninja; extra == "runtime-common"
 Requires-Dist: orjson; extra == "runtime-common"
 Requires-Dist: packaging; extra == "runtime-common"
 Requires-Dist: pillow; extra == "runtime-common"
@@ -233,23 +236,20 @@ Requires-Dist: pydantic; extra == "runtime-common"
 Requires-Dist: python-multipart; extra == "runtime-common"
 Requires-Dist: pyzmq>=25.1.2; extra == "runtime-common"
 Requires-Dist: torchao>=0.7.0; extra == "runtime-common"
+Requires-Dist: transformers==4.48.3; extra == "runtime-common"
 Requires-Dist: uvicorn; extra == "runtime-common"
 Requires-Dist: uvloop; extra == "runtime-common"
-Requires-Dist: xgrammar==0.1.14; extra == "runtime-common"
-Requires-Dist: ninja; extra == "runtime-common"
-Requires-Dist: transformers==4.48.3; extra == "runtime-common"
-Requires-Dist: llguidance>=0.6.15; extra == "runtime-common"
+Requires-Dist: xgrammar==0.1.15; extra == "runtime-common"
 Provides-Extra: srt
 Requires-Dist: sglang[runtime_common]; extra == "srt"
-Requires-Dist: sgl-kernel==0.0.3.post6; extra == "srt"
-Requires-Dist: flashinfer_python==0.2.2.post1; extra == "srt"
+Requires-Dist: sgl-kernel==0.0.5; extra == "srt"
+Requires-Dist: flashinfer_python==0.2.3; extra == "srt"
 Requires-Dist: torch==2.5.1; extra == "srt"
 Requires-Dist: vllm<=0.7.2,>=0.6.4.post1; extra == "srt"
 Requires-Dist: cuda-python; extra == "srt"
 Requires-Dist: outlines<=0.1.11,>=0.0.44; extra == "srt"
 Provides-Extra: srt-hip
 Requires-Dist: sglang[runtime_common]; extra == "srt-hip"
-Requires-Dist: sgl-kernel==0.0.3.post6; extra == "srt-hip"
 Requires-Dist: torch; extra == "srt-hip"
 Requires-Dist: vllm==0.6.7.dev2; extra == "srt-hip"
 Requires-Dist: outlines==0.1.11; extra == "srt-hip"