sglang 0.5.1__py3-none-any.whl → 0.5.1.post1__py3-none-any.whl

sglang/srt/layers/quantization/deep_gemm_wrapper/configurer.py

@@ -1,5 +1,7 @@
 import logging
 
+import torch
+
 from sglang.srt.utils import get_bool_env_var, get_device_sm
 
 logger = logging.getLogger(__name__)
@@ -7,8 +9,10 @@ logger = logging.getLogger(__name__)
 
 def _compute_enable_deep_gemm():
     sm_version = get_device_sm()
-    # TODO fix blackwell fp8
-    if sm_version != 90:
+    if sm_version < 90:
+        return False
+    # TODO fix deepgemm cu129 fp8 issue
+    if torch.version.cuda == "12.9":
         return False
 
     try:

sglang/srt/models/grok.py

@@ -842,10 +842,6 @@ class Grok1ForCausalLM(nn.Module):
         if self.is_weights_presharded:
             setattr(DefaultModelLoader, "_prepare_weights", _prepare_presharded_weights)
 
-        default_replicate_lm_head = False
-        self.replicate_lm_head = getattr(
-            config, "replicate_lm_head", default_replicate_lm_head
-        )
         self.replicate_embedding = getattr(config, "replicate_embedding", False)
 
         self.model = Grok1Model(

sglang/version.py

@@ -1 +1 @@
-__version__ = "0.5.1"
+__version__ = "0.5.1.post1"

{sglang-0.5.1.dist-info → sglang-0.5.1.post1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sglang
-Version: 0.5.1
+Version: 0.5.1.post1
 Summary: SGLang is yet another fast serving framework for large language models and vision language models.
 License:                                  Apache License
                                    Version 2.0, January 2004
@@ -374,7 +374,7 @@ 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. [Register here](https://lu.ma/gbfhjvuo).
+- [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 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/)).

{sglang-0.5.1.dist-info → sglang-0.5.1.post1.dist-info}/RECORD

@@ -9,7 +9,7 @@ sglang/global_config.py,sha256=ZMTux_PsGnvkyJ0kTFwhTdbnFwIjjpGDogut_9Lu4Vo,1732
 sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
 sglang/profiler.py,sha256=tEHzHerXC-ymk4OrkoUcMbgcGHmb8VESthsNSP2Yx9w,4417
 sglang/utils.py,sha256=dC2PNkKYTgDHsNrWdZJ74GvaXGSHCeIk_aZ-TA89OhY,16380
-sglang/version.py,sha256=eZ1bOun1DDVV0YLOBW4wj2FP1ajReLjbIrGmzN7ASBw,22
+sglang/version.py,sha256=cOaLNeTwkPXJFU_tSZi-530rXF3qjxypENpk5W_HqGE,28
 sglang/eval/llama3_eval.py,sha256=mLNRZJIqV4CfqrY8UGnJEcHw2Xsyr1eyYZgFSUFYr1g,9997
 sglang/eval/loogle_eval.py,sha256=-CC2s2kh5qUoDrHRkQVkC_jNvBgNojXbf456ny5s78s,4557
 sglang/lang/api.py,sha256=rcp3GeoyZhmJ0GDLPRkuZNcxd0TBJy_wfUDpcmQoqW8,7210
@@ -614,7 +614,7 @@ sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a
 "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FFBjSWlpKXMxfAUUYUqXbOK_Hd7qBeBsfbcaa9uB4qY,3249
 sglang/srt/layers/quantization/deep_gemm_wrapper/__init__.py,sha256=esJMd0Yuj68t6QYOpmIFuiWP2J2dxTMC4bRBNH0Xk6I,26
 sglang/srt/layers/quantization/deep_gemm_wrapper/compile_utils.py,sha256=FaHKgY9y1oi3zlCmkorSK4r4dhPy7NiviVIhtabvzjM,11536
-sglang/srt/layers/quantization/deep_gemm_wrapper/configurer.py,sha256=ajxS0OUYp2RbHtP3dB71EzsXKQEHMwP4Wo7-pkJZ9u4,771
+sglang/srt/layers/quantization/deep_gemm_wrapper/configurer.py,sha256=LPO728y32uHQOGfk8U26jDIbGlAcQdD-ZYaqfboyhzM,853
 sglang/srt/layers/quantization/deep_gemm_wrapper/entrypoint.py,sha256=On_uvP1ThxgQFAzOWKoteM0vIOHqmtI127A4qbPtmbc,3264
 sglang/srt/layers/quantization/quark/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sglang/srt/layers/quantization/quark/quark.py,sha256=EqQvWzAc6nd_yl6UIMIwmozKif4I8fRNfBum040UUEQ,14613
@@ -728,7 +728,7 @@ sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtY
 sglang/srt/models/gpt_oss.py,sha256=qXa6BAm8gBTuCSX0ax3Iz4SkIIko0s1u8sVu8kJVzkY,44930
 sglang/srt/models/granite.py,sha256=8q92shxVPAp_cJDohJATffSGd7Z0Oi-vF5jpY6DlK4s,19840
 sglang/srt/models/granitemoe.py,sha256=j1rgZ62CbBioECjUblDCw_NneDQgY_QJODsI0fqXVO8,13779
-sglang/srt/models/grok.py,sha256=Va-kPr1XxjK4jWhVkvqzRl3e6OWuGZnbhp1TadadkeQ,40679
+sglang/srt/models/grok.py,sha256=8KCR13LtdnhswJrIwgb0sdFW0OxCA8GQ0fbN8gb856E,40518
 sglang/srt/models/hunyuan.py,sha256=RSZErx-Swt65kmfvSXJQJJ6KhjucuNY4UUVLWC8hWaA,31102
 sglang/srt/models/idefics2.py,sha256=U3khd3hbdawJeRNXsxmaKHdssOCT5TPOZ1D-2_zHoQo,12079
 sglang/srt/models/internlm2.py,sha256=F_iNY1gUqzAjAuUatcE47gnrcoTh5_08PY2Rw9tKr9M,13150
@@ -862,8 +862,8 @@ sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqn
 sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
 sglang/test/attention/test_prefix_chunk_info.py,sha256=hpoDe2wfSa6RlUbfyri_c0iyBTb35UXGL9I2Xh6jamM,7772
 sglang/test/attention/test_trtllm_mla_backend.py,sha256=ilrX20-1hVo0nmKN0GTqzKoUmDarTXGS8nWJ6c8RHVY,41792
-sglang-0.5.1.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
-sglang-0.5.1.dist-info/METADATA,sha256=ay7jXrW9PA-uzilMp3qXaFmXLVietplAqnH1Buh4l8c,27609
-sglang-0.5.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-sglang-0.5.1.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
-sglang-0.5.1.dist-info/RECORD,,
+sglang-0.5.1.post1.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
+sglang-0.5.1.post1.dist-info/METADATA,sha256=IgpodN0t3cCFIUqlTb5l0iNjc0obANmUVH_Wv3j5i8M,27803
+sglang-0.5.1.post1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+sglang-0.5.1.post1.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
+sglang-0.5.1.post1.dist-info/RECORD,,