sglang 0.4.0.post1__py3-none-any.whl → 0.4.0.post2__py3-none-any.whl

sglang/test/test_utils.py

@@ -568,6 +568,7 @@ def run_bench_serving(
         disable_tqdm=False,
         disable_stream=disable_stream,
         disable_ignore_eos=False,
+        return_logprob=False,
         lora_name=None,
         extra_request_body=None,
         profile=None,
@@ -719,13 +720,13 @@ def run_and_check_memory_leak(
 
     # Clean up everything
     kill_process_tree(process.pid)
-    kill_process_tree(process.pid)
     stdout.close()
     stderr.close()
     if os.path.exists(STDOUT_FILENAME):
         os.remove(STDOUT_FILENAME)
     if os.path.exists(STDERR_FILENAME):
         os.remove(STDERR_FILENAME)
+    kill_process_tree(process.pid)
     t.join()
 
     # Assert success
@@ -733,7 +734,7 @@ def run_and_check_memory_leak(
     has_leak = False
     has_abort = False
     for line in output_lines:
-        if "The server is fired" in line:
+        if "Uvicorn running" in line:
             has_new_server = True
         if "leak" in line:
             has_leak = True

sglang/utils.py

@@ -1,4 +1,4 @@
-"""Common utilities."""
+"""Common utilities"""
 
 import base64
 import gc
@@ -79,7 +79,14 @@ class HttpResponse:
         return self.resp.status
 
 
-def http_request(url, json=None, stream=False, api_key=None, verify=None):
+def http_request(
+    url,
+    json=None,
+    stream=False,
+    api_key=None,
+    verify=None,
+    method: Optional[str] = None,
+):
     """A faster version of requests.post with low-level urllib API."""
     headers = {"Content-Type": "application/json; charset=utf-8"}
 
@@ -90,7 +97,7 @@ def http_request(url, json=None, stream=False, api_key=None, verify=None):
     if stream:
         return requests.post(url, json=json, stream=True, headers=headers)
     else:
-        req = urllib.request.Request(url, headers=headers)
+        req = urllib.request.Request(url, headers=headers, method=method)
         if json is None:
             data = None
         else:

sglang/version.py

@@ -1 +1 @@
-__version__ = "0.4.0.post1"
+__version__ = "0.4.0.post2"

{sglang-0.4.0.post1.dist-info → sglang-0.4.0.post2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sglang
-Version: 0.4.0.post1
+Version: 0.4.0.post2
 Summary: SGLang is yet another fast serving framework for large language models and vision language models.
 License:                                  Apache License
                                    Version 2.0, January 2004
@@ -215,6 +215,7 @@ 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: decord; extra == "runtime-common"
@@ -232,16 +233,17 @@ Requires-Dist: psutil; extra == "runtime-common"
 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; extra == "runtime-common"
+Requires-Dist: torchao>=0.7.0; extra == "runtime-common"
+Requires-Dist: gemlite; extra == "runtime-common"
 Requires-Dist: uvicorn; extra == "runtime-common"
 Requires-Dist: uvloop; extra == "runtime-common"
-Requires-Dist: xgrammar>=0.1.4; extra == "runtime-common"
+Requires-Dist: xgrammar>=0.1.6; extra == "runtime-common"
 Provides-Extra: srt
 Requires-Dist: sglang[runtime_common]; extra == "srt"
 Requires-Dist: torch; extra == "srt"
 Requires-Dist: vllm<=0.6.4.post1,>=0.6.3.post1; extra == "srt"
 Requires-Dist: cuda-python; extra == "srt"
-Requires-Dist: flashinfer>=0.1.6; extra == "srt"
+Requires-Dist: flashinfer==0.1.6; extra == "srt"
 Provides-Extra: srt-hip
 Requires-Dist: sglang[runtime_common]; extra == "srt-hip"
 Requires-Dist: torch; extra == "srt-hip"
@@ -311,8 +313,11 @@ Requires-Dist: sglang[test]; extra == "dev-hpu"
 
 --------------------------------------------------------------------------------
 
-| [**Blog**](https://lmsys.org/blog/2024-07-25-sglang-llama3/) | [**Documentation**](https://sgl-project.github.io/) | [**Join Slack**](https://join.slack.com/t/sgl-fru7574/shared_invite/zt-2tmmp6flg-89dOlJW2TjnBrTRk1I_~GA) |
-[**Join Bi-Weekly Development Meeting**](https://docs.google.com/document/d/1xEow4eIM152xNcRxqZz9VEcOiTQo8-CEuuQ5qTmkt-E/edit?usp=sharing) | [**Slides**](https://github.com/sgl-project/sgl-learning-materials?tab=readme-ov-file#slides) |
+| [**Blog**](https://lmsys.org/blog/2024-07-25-sglang-llama3/)
+| [**Documentation**](https://sgl-project.github.io/)
+| [**Join Slack**](https://join.slack.com/t/sgl-fru7574/shared_invite/zt-2tmmp6flg-89dOlJW2TjnBrTRk1I_~GA)
+| [**Join Bi-Weekly Development Meeting**](https://docs.google.com/document/d/1xEow4eIM152xNcRxqZz9VEcOiTQo8-CEuuQ5qTmkt-E/edit?usp=sharing)
+| [**Slides**](https://github.com/sgl-project/sgl-learning-materials?tab=readme-ov-file#slides) |
 
 ## News
 - [2024/12] 🔥 SGLang v0.4: Zero-Overhead Batch Scheduler, Cache-Aware Load Balancer, Faster Structured Outputs ([blog](https://lmsys.org/blog/2024-12-04-sglang-v0-4/)).

{sglang-0.4.0.post1.dist-info → sglang-0.4.0.post2.dist-info}/RECORD

@@ -1,18 +1,19 @@
 sglang/__init__.py,sha256=b2oIdWzp5P8SzieeOs2TzJoN3Do3tfJbV8gZS_imVcs,1619
 sglang/api.py,sha256=NdO6cYnklnEBQBKqQjlqI8-P1EownKQ71t5ibCGhEVo,6953
 sglang/bench_latency.py,sha256=oZjSAzX7dUiSu-zdz0dkyUPo-qAX_lsXFH1gf03akgI,76
-sglang/bench_offline_throughput.py,sha256=3OrFI26PmoVTU3pQrBFC50AZI7HpKKuk4vYycbkDjhY,12428
-sglang/bench_one_batch.py,sha256=vxXSCQRTMeJUtJKsSoP6tLdoWTdFp1mhwsLpKHccs2c,15858
+sglang/bench_offline_throughput.py,sha256=rgMWDhA1Hai0gKBzxc0dzTWfI8l39Cyw2VOCyMt1YyY,12771
+sglang/bench_one_batch.py,sha256=aF0onHeRjy7AYVjsq1IA3rZEhUuYXuslg1fAhuvJ2yo,16120
 sglang/bench_one_batch_server.py,sha256=-fV9FTLNNcSIy0pgYeggXedPVK0fVsXZqVQswT8OMOY,5945
-sglang/bench_serving.py,sha256=Oa_Qi7YApv37jGDAmuIaZSIhayvRpKq9GZGZLXBU-9I,52924
-sglang/check_env.py,sha256=q1sdYL-gcKSCeIZMk7sUMh9rjM71f-EUgp07OGPSbZM,5446
+sglang/bench_serving.py,sha256=zv_EcbWno79j7WYFL2m6BfCLT6iSOfGV4uwGbDg9KQA,53141
+sglang/check_env.py,sha256=4OqpZaEJOfBM6-vtPILto5kqDmgiZM1Koc7lK78A7CI,8427
 sglang/global_config.py,sha256=fnT0U9vlHdGaQFKN9tYTnUF4-eVW4HYQURd5zvPtrg0,1286
 sglang/launch_server.py,sha256=4y2QeSj0wVNB9MJQZeahD4ahTDU6gwqo7MPUytyFop0,403
 sglang/launch_server_llavavid.py,sha256=tGc17S1vUfLwbi1GB26oOdXxTWr7gjlqpTrPnrMRNO8,1007
-sglang/utils.py,sha256=r4Dw-xffcrTRposls-gqyoYxjgJNYhVduK_6bDN_Vj4,11526
-sglang/version.py,sha256=7uCJCJaoOPp2kM9G7GksnnqBLasAAF_4Hm_M-3jPgmc,28
+sglang/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
+sglang/utils.py,sha256=23jf4Mz8E5p5a6JOkjnfYZixdjZUk88F_mZ8rZcby5Q,11597
+sglang/version.py,sha256=OUNovuQ1RrdJFYetl0e0U0556H_wiyjhVks9-l-zF94,28
 sglang/lang/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-sglang/lang/chat_template.py,sha256=jprS3-In2FTUoedKwZg-HYvDwU8RTIYntOlf2zoN2sU,14814
+sglang/lang/chat_template.py,sha256=cnfjjxIIcYRGRxXlJlOGnpFxFuhMHut7DS52LsOMKcA,15826
 sglang/lang/choices.py,sha256=-W1DVw9N9ZliVpvmWrzIXG4cswAah8eMQrHWzkS3D8o,6234
 sglang/lang/compiler.py,sha256=o1C6G3TzhjSlsH-doTPy5oiVehr57dxNTa5oZw5TTAI,7639
 sglang/lang/interpreter.py,sha256=SBjejhLhTKzNM0HbjtTg5r17WPJ64WFSk6lcM_SCWKs,30717
@@ -23,16 +24,16 @@ sglang/lang/backend/anthropic.py,sha256=EXRX7xJgA5KZszX7toSLVnKzFQ5EO0Loj-YjHFtx
 sglang/lang/backend/base_backend.py,sha256=tdoh9YF3CyekY1BKiX9n7-aA4srDWIuA4RDJLM7q8qg,1985
 sglang/lang/backend/litellm.py,sha256=ugmL7sfUxkUHVbHtwNzHgdQAEd4UCjNQboFuE3KThcY,2450
 sglang/lang/backend/openai.py,sha256=qM7eVH_kMxnDd2rpxOH0v76KxtOJFlAwgLgWIKvFGCI,15060
-sglang/lang/backend/runtime_endpoint.py,sha256=IWbrAKrUkzNOvwV6V9_y6pkTr2SUYEkKBT-3kirgad0,10514
+sglang/lang/backend/runtime_endpoint.py,sha256=dfs-yZ1ekKmnbpZLluQHWPmMeZJKbaaZRRGYRa9eBE8,10541
 sglang/lang/backend/vertexai.py,sha256=O-iBLD-y3vq80UxnrAoJri7bxpgd-_eakZ88Cf8bEGA,4855
 sglang/srt/_custom_ops.py,sha256=Y4gyTDGhWz-W2Igq25Ojm8XFiyvkawW9I-79iwYvxJ0,3574
 sglang/srt/conversation.py,sha256=u9zFU8aMYzwHUbQRKU76B_T-jfLlPoxUcWG_nRbDM2I,21201
 sglang/srt/hf_transformers_utils.py,sha256=38Ms0H2-VMerOS6jnczcFtZMS6lhw9B5rSWKAfxVUfQ,7945
 sglang/srt/mm_utils.py,sha256=1ScBunw_x4W8ebM_AcJ62-1T2mfT8NlMJqdAhkF1lb0,12367
-sglang/srt/model_parallel.py,sha256=1QmoJGzflbYVQ9OpG6YqmtQ6T0QI5HRl61o9nkhvu7o,3570
-sglang/srt/server.py,sha256=Ne7yfXUIST72C0u_f-KoSKLT9PbE9vXtULRa0I_qJzw,34620
-sglang/srt/server_args.py,sha256=TYesukaMQUnv0ru1mY6quU8S1UegxRL0eOCVdaNMyNY,34302
-sglang/srt/utils.py,sha256=F1gffBAjmtLpvvy60Ex38emttpjr0Xn7G1Vj_pzMxtA,42197
+sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
+sglang/srt/server.py,sha256=tEciMH_U6WIZYPUGDDM0c4BQ16cvgVdA4II-ksPZoMo,34621
+sglang/srt/server_args.py,sha256=LgnQ-kBJZ3E7hMMZj9bSK0mn7Bhjk1nJHxLcxl-lGTM,34572
+sglang/srt/utils.py,sha256=WWEcMJHmvlOjiqE9UicT0ZYwa2PUKDZorAk2Y8PPRBI,42039
 sglang/srt/configs/__init__.py,sha256=_usVIXHQjft4PAJ1Y-yGQOn2QNOv501GYMlQwpGXbns,208
 sglang/srt/configs/device_config.py,sha256=dResqHjkg_dq10v6rnVpbXpvABZRB0jylOm-2_JAnx0,428
 sglang/srt/configs/exaone.py,sha256=Duxd4yQoKy8GWEzZD_kCY_OzmN_67CTJL_Kgn0eXk3g,10731
@@ -43,7 +44,7 @@ sglang/srt/constrained/__init__.py,sha256=UWZNVLvOT5ZBX8M36sONgDmnKtkQ0cSfhQD2jO
 sglang/srt/constrained/base_grammar_backend.py,sha256=FhVm7PxhXDl0joV9NP5RjKgz7dR1dZvUAQnh0mdtvVY,2353
 sglang/srt/constrained/outlines_backend.py,sha256=CipNHNNXs8xtnJNVNe6FCwZUlSbIXbGmWVlZz3hUpFQ,6820
 sglang/srt/constrained/outlines_jump_forward.py,sha256=iZWXeR3gNYoMubLGyFmLPO4V2YsN5DiGjD71Xk9iFaE,6418
-sglang/srt/constrained/xgrammar_backend.py,sha256=qk9tzkoJnp0yDBVdIsB5YIL8PLEF_d8ZsSsv5HQAglo,4471
+sglang/srt/constrained/xgrammar_backend.py,sha256=4It9_GqU4UZFhxIw_7hkzpXaMPUtksk6Xfe0Agsfw7A,4620
 sglang/srt/distributed/__init__.py,sha256=__tl9Frrf3PFrSyNYcn5i-y2rL-J4-Qn6RJwrsZ4xgc,83
 sglang/srt/distributed/communication_op.py,sha256=ZoIhboZyefiAwr-1K-wF3rAFSQ4Wt-RxXpsX443Gbt4,1157
 sglang/srt/distributed/parallel_state.py,sha256=HplRH5S0AWdwSdhoHYX9_UWQZlFjh2Z1LHaz68EXlpE,47555
@@ -62,56 +63,56 @@ sglang/srt/layers/custom_op_util.py,sha256=0vu-yX2wwonmO1L_o5G7SA6C-8XuhDIh9rPDv
 sglang/srt/layers/fused_moe_patch.py,sha256=DMIyrwOON7OSidKZdreL5HzMhP0AD5Ues0xdY-ADOQw,4471
 sglang/srt/layers/layernorm.py,sha256=nRQ1w1xSUcU-zlqVC61BnGG6otS5W1w9VaSzeXizrx4,4037
 sglang/srt/layers/linear.py,sha256=dF2HvqiMbhWlCjvkLFRCcgUFGhG-B0keM_CIpjvgTtg,46154
-sglang/srt/layers/logits_processor.py,sha256=oZNu9pNNgmswhuw8irlLm0SfpVrD7cFf-GdfPsLZGHE,13227
+sglang/srt/layers/logits_processor.py,sha256=JlOU0x8vBGIuTwHSdjR6Kly9_uzilBMv0NE_rvUx0W4,14747
 sglang/srt/layers/pooler.py,sha256=rj2lygvleBnyLCBZ8I11HGMgpfIDsT0l3PIkshJwdu4,1606
-sglang/srt/layers/radix_attention.py,sha256=tqC6bqQYqKpdWLoL9j8jmZzgBAMX0XKcKaPlgfJ0uhc,2115
+sglang/srt/layers/radix_attention.py,sha256=E4cmvkcCdCtb6VyLNrCKy1D6VwHQ063oH3JQXPaRy6w,2178
 sglang/srt/layers/rotary_embedding.py,sha256=29tx3JNR40AoXqBa2cFGBjva9vU2xgFipETlpMaaZas,3985
-sglang/srt/layers/sampler.py,sha256=yEnAKBFpRcz09Ro1DTQ_C5lUBsSko3DR1jxAFGvnaLU,4677
-sglang/srt/layers/torchao_utils.py,sha256=I1-sUDvfuSymfBk33WpIqz4gHlurh_iey1tO5T23Gv8,2499
+sglang/srt/layers/sampler.py,sha256=k4Op_HMkQfT7t9wgQwBVotfTUXEocrzRyQqEFnff1pc,5511
+sglang/srt/layers/torchao_utils.py,sha256=07Fe2Csdh1JiQKPGGHWkbq0-a6bV7Cq136ygdtVAhgI,3708
 sglang/srt/layers/vocab_parallel_embedding.py,sha256=slGwLiWjuFLCUdRe-GTlfumyZpqVX9VF6No_UGOT-hA,21624
 sglang/srt/layers/attention/__init__.py,sha256=KIJhzOJWYioQE7Va4D83-V-ZUZVMZcczuNgDC3dlSRo,2583
 sglang/srt/layers/attention/double_sparsity_backend.py,sha256=RQdEKRykSLf9ilnaHmR6T7RFqh4emH_adfB3aJN2BUU,10920
-sglang/srt/layers/attention/flashinfer_backend.py,sha256=SqwnXV4hstxN-OUev5ZIACHbXwNRHLM7NK85LB-259A,24855
+sglang/srt/layers/attention/flashinfer_backend.py,sha256=NgeigL1WiPOuOry0Gbxv-6HEcERB8Du0mBJgYcTVIAA,24943
 sglang/srt/layers/attention/torch_native_backend.py,sha256=nQdeqWEMMH_wrod5wssDCJG-uPKm0uslvkALKqPRPQ8,10509
-sglang/srt/layers/attention/triton_backend.py,sha256=HEnmQ14eOsnDkTWhx3r-02PxQfNFDMkGLDXATbEquGE,6752
-sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=BE63WhKiutSNkhJLsRwvfsRy-ExvuAv7FZyoWv73ul8,18744
+sglang/srt/layers/attention/triton_backend.py,sha256=-TobyZHwlbJ5HhbFg-jgCqVOw4Y-opgEuFo-EusASQc,6264
+sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=oJ_UK1t229zF3hbTDiQe7t-X-IbM2dOxx4U2ch-vmjA,17847
 sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=1pSXfY3EEaM7iRN_uElHnAfsrJMhTFbu9fj8Z0O2PbE,21480
 sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=tZJhzqcf1KKMT8z7_32eVk_D1NHP71c-S3UNxemfAHM,11542
 sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=lojFXRZMLWkzS2Y8uxaolnQhXaWKG19mCAWaF5KQeiI,6087
 sglang/srt/layers/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sglang/srt/layers/ep_moe/kernels.py,sha256=wb_S2qLxoWWgQu9coXy0XLNGvHzdZSdwXr0PGy4QySg,10940
-sglang/srt/layers/ep_moe/layer.py,sha256=_z_oVTIbHXxz_r7ne0TZdafMn0cdm0mgjQk9_UT3Rwk,22845
+sglang/srt/layers/ep_moe/layer.py,sha256=uMropMhU-MaycoxSLxcfD0jZC_cuL_boRbIu86mbZjY,23034
 sglang/srt/layers/fused_moe_triton/__init__.py,sha256=PHKFqd2hPOO-g9kSMseg2g76lpg9OGXQDThWU6bt9vs,902
-sglang/srt/layers/fused_moe_triton/fused_moe.py,sha256=qwfRBOeY5DT48Q6z71Eh9cjFehvs_K6eLIVWNL044Ug,28363
-sglang/srt/layers/fused_moe_triton/layer.py,sha256=URDkTt8xEqnqpO5tb_3L7JlhlO53VWfqDDNSRYEu-LY,21545
-sglang/srt/layers/quantization/__init__.py,sha256=Y0rZz3rj6Dg5_QDvsXJd7cZ9NWFFngYrBMA8LuW4WTk,5943
+sglang/srt/layers/fused_moe_triton/fused_moe.py,sha256=fLGmkY6imJYjEw9-3-jJthkMcFGMBcu9HCNIuxAzMhE,29625
+sglang/srt/layers/fused_moe_triton/layer.py,sha256=eMpbZlP3FAQxbHochis7ybZ-fsNBP0PzKF1PN0Xo7so,21517
+sglang/srt/layers/quantization/__init__.py,sha256=FgNy_zNWMWnq3lEGyCSyfLSQtcZtWlq99JilkmEDW7I,4594
 sglang/srt/layers/quantization/base_config.py,sha256=daK9p0aijMszLUm1W4Pc33FK87MdqYK1NoWFKif-j80,4599
-sglang/srt/layers/quantization/fp8.py,sha256=uyX8U-5cEDLq2XWuG5fX6GKHBla-Z188vi5dgX1dLNk,21975
+sglang/srt/layers/quantization/fp8.py,sha256=3oIUPaD0PBXQyTKr44I0YJ8XXDdwyoS_-ZA97XdSxXE,24143
 sglang/srt/layers/quantization/fp8_utils.py,sha256=eJDLLDu8ZbrbE3BfFIf89JlIMPOP-14DesbeVsajW0Q,1035
 sglang/srt/lora/lora.py,sha256=-o2mBmUvoVpdkgdAkWTARN4kfyep3UNEJLcg6moh0SU,15056
 sglang/srt/lora/lora_config.py,sha256=a2fTQESlCbG1xLiBYy4ptZ6c0Burcqyg1_6V1XSok-Y,1506
 sglang/srt/lora/lora_manager.py,sha256=DHiqdl0_4wQ5PxZBZtlCpP14515mDV2_H9tzL3Rdss8,12886
 sglang/srt/managers/data_parallel_controller.py,sha256=psI4FAuBGjtdnEuwagnGdtRqvqSSxOROfNKQqVDqlVA,8382
-sglang/srt/managers/detokenizer_manager.py,sha256=TtrtE37XT5XcJzk8-R5rHZ16NHTPd5XZi8hf3h-sB2A,7462
+sglang/srt/managers/detokenizer_manager.py,sha256=nZkbwt4yty_oy8rvg4T7PbgyVLoBLohvHl25xlQpBoo,8439
 sglang/srt/managers/image_processor.py,sha256=Y8RgyrzbJjJTpjbnZDa5qiiG5wWjZ68rOXUPDi6kkFo,13698
-sglang/srt/managers/io_struct.py,sha256=d_kctmHcNBzzaP5lUEIpdrVVsob4dNOetMHkobUJZz4,14439
-sglang/srt/managers/schedule_batch.py,sha256=Zf45mPuz7YyYSSuOXikdjnRfOEG5FE1JKF9OftWxKdw,45045
-sglang/srt/managers/schedule_policy.py,sha256=7QuIsJDRzkrvs3IJk10oOfL4Me0UZwDYvRniT1fSFuo,12620
-sglang/srt/managers/scheduler.py,sha256=NpBQZjTCysFGZdXvnWG4jqd_eiWrV0a2EUtMpEItDcU,59650
+sglang/srt/managers/io_struct.py,sha256=_LWWqT3LNwZGaWhg2d3kTg1V2MTHKzRasCvxF9Nfpi4,15429
+sglang/srt/managers/schedule_batch.py,sha256=SAd7sxhoC3Bp8_xd-TEcXEFZBlGZPbn8-wMvBcjU55Q,45607
+sglang/srt/managers/schedule_policy.py,sha256=cLNi__smbg02keWgUMfB_nEM3vllocPB0XyG1P5qO7I,15469
+sglang/srt/managers/scheduler.py,sha256=QlcVMtrLlNcBOkVISdO556jrK8a4LE4ULskC0oCH2IQ,61776
 sglang/srt/managers/session_controller.py,sha256=Yp-IV3rXczACZxZXmF-QxW9CWICGy8KHQ9ttBGJ8WXA,2800
-sglang/srt/managers/tokenizer_manager.py,sha256=XPaSXB6b23u95viFqlqd-tdyrNMMOOiSDWviz_g7UBM,29890
+sglang/srt/managers/tokenizer_manager.py,sha256=gnCCdB5XDobOoBKptwv-o0yYqDkMUxL78s0zBno5lM4,31219
 sglang/srt/managers/tp_worker.py,sha256=X1EwFX3FSsmXx7jeeX2tjZRocaujabQYWm-M-0CFEBE,7363
 sglang/srt/managers/tp_worker_overlap_thread.py,sha256=-QNBJRKxraa9Xt2WI1AFzZYdneIJ1eXv0GjFzDqXoE0,8926
-sglang/srt/mem_cache/base_prefix_cache.py,sha256=qEQwEkG4E5rab2ZoTqcesf5pR_J4nV2jBxIHsBJHtIM,924
-sglang/srt/mem_cache/chunk_cache.py,sha256=VcCpyrf5FOQ5xoKeOouCI5ZQLkZo_pgY1SPbDDkagGg,2492
+sglang/srt/mem_cache/base_prefix_cache.py,sha256=QC8HS8RC5DXu14kyXsxAgEUsn0f932p2DjqzbKjc6Bs,962
+sglang/srt/mem_cache/chunk_cache.py,sha256=R2gHAuqKd5ayQW3NnsgoGUH31---Z5izCDyCqLL0FjQ,2524
 sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
 sglang/srt/mem_cache/memory_pool.py,sha256=l9_srwXEfIIDF46nxykbHIOo1VSvU5_Ew3H0r5EC7Fo,11072
-sglang/srt/mem_cache/radix_cache.py,sha256=DzLCO_gYQ7X_C2NJSEHzzMZhb5HzWjKF9wXJQsnzr8M,10427
+sglang/srt/mem_cache/radix_cache.py,sha256=c5voySV5L855c0G9cBEc9iQ4nR7PDDmg0V6fWWJHcq4,10945
 sglang/srt/metrics/collector.py,sha256=ZWoFx_FKN0sNMSZ8RJWUVQ0RFEYhIHxdw0d4TZTluMU,6861
 sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
-sglang/srt/model_executor/cuda_graph_runner.py,sha256=6llBpL1Zdb_iNKmcA5yL_ieR0-hP1JSNmUruTiOkiS8,15796
+sglang/srt/model_executor/cuda_graph_runner.py,sha256=kZ3nV03MD8EQYQB38u4_88_wyW4unECxAdMVICpPyuk,16241
 sglang/srt/model_executor/forward_batch_info.py,sha256=L5mVoW5SaO6To-7nGk0TZM-FFB5_78cARpJ-aC2rwD0,12883
-sglang/srt/model_executor/model_runner.py,sha256=cO4KZfaNtbpeDwzNrnHwtyg2UG7iFl7MW6M-sXl-lO8,29523
+sglang/srt/model_executor/model_runner.py,sha256=MLYBcYIQihu2I3PBTUghiU2mSWsDMzlKzcnX7yHa9JU,29837
 sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
 sglang/srt/model_loader/loader.py,sha256=VBrY4W9CiVvS_D8yXhdkW9jReV9rSMSkJplabz0Fxgk,43528
 sglang/srt/model_loader/utils.py,sha256=0NaMR67fESFopaklmsleiL27XH1QUrjZW246MUu1EJ0,1369
@@ -124,18 +125,19 @@ sglang/srt/models/deepseek.py,sha256=BVNICGoLjQoHmR5lc31YrZ6YbxSRTBilHqlLsALr2u8
 sglang/srt/models/deepseek_v2.py,sha256=YKSrqagVcSUwCAi-rwIph-Xu12GrNETMNKxgnffWod8,35349
 sglang/srt/models/exaone.py,sha256=dkERTZVxrRroqu5AGLP7D4N6n8HvDqlNaDQUIe15mZY,13038
 sglang/srt/models/gemma.py,sha256=ydRqsG-7004r1fAiz01LHUmcj_6XN0Tn4xO1keJnMQk,12126
-sglang/srt/models/gemma2.py,sha256=vPrAasJajitQHB9ZqMFut58xNsOm3fk2m05a-feQL10,14600
-sglang/srt/models/gemma2_reward.py,sha256=hJw0hXNPyQSpazkVJVYiW04OtTZH0GiLI-JJef_kaGs,2529
+sglang/srt/models/gemma2.py,sha256=41PlW8pMb4rMETdAni_JWDhZeIn_QsTQireAyUjsURA,15848
+sglang/srt/models/gemma2_reward.py,sha256=nJ01KfqLSJtqMLm3sG8p2mGZFK1xhhjh7I7Ccb-_Hq8,2494
 sglang/srt/models/gpt2.py,sha256=2je1kE09sGcaORWnJuGYAkcwwOrT9EK-KhQaoCKjCSA,9517
 sglang/srt/models/gpt_bigcode.py,sha256=tovyOdJu2x3LkzmkdFXX_iJdkxuyChIDxwgvPBy6UPo,9528
-sglang/srt/models/grok.py,sha256=68uS5wuBCc0ClMT2iQl4iGPNbxASLz6c7dCYFrLxGaM,13661
+sglang/srt/models/granite.py,sha256=AeQY9Dxd1ZnwgCYBK0vSXXiMGM-yt9iaOVf_ruOUHXw,20409
+sglang/srt/models/grok.py,sha256=UWvVEYfEoH0jGNFSbXpO66OGW5pzmIHlNKcn9gRZEoQ,15664
 sglang/srt/models/internlm2.py,sha256=_xcKtd6YtEFUTozaN-yUb0xbSYckRpomfPSKcAk4j-Y,12127
 sglang/srt/models/internlm2_reward.py,sha256=8K26A9oIFFGx_9U2mF87j7FX8K87HGKMnVL3ht1Uc7I,2398
-sglang/srt/models/llama.py,sha256=elHH1x3nS_c13Tgx97ExXvbQ9Gzwqg5JCjyQ9qNzmF4,19269
-sglang/srt/models/llama_classification.py,sha256=EdXmiMyfJ9NH5P-Wel7SRhf_v8ddFFhVJMQgzDt0oVk,3377
+sglang/srt/models/llama.py,sha256=S7nS05hhFGghXu0v-w9RZyBTY6OCEVF5Aaw4GX_E_9g,19929
+sglang/srt/models/llama_classification.py,sha256=DwboM1xHXdf3Fddf7xGnrfdOLJwXdiJs994cIpAPa2g,2984
 sglang/srt/models/llama_embedding.py,sha256=rh-AiczPY_pTpzcACHvSMVjh1hsV_MZBBwP0LQxPsGM,3130
-sglang/srt/models/llama_reward.py,sha256=JVaiTK4gVXNMimeq3kKkv7dt5Hc77hPqF4ewvmzjJes,4622
-sglang/srt/models/llava.py,sha256=l9mqS9wl_l6ARC-K1UUe7XsB5k9sZratMNQEwx5IjR0,25229
+sglang/srt/models/llama_reward.py,sha256=oPxh5E2UkxLULNdR68dFvt2I7j33CJFN6nyA-8L2_cg,4516
+sglang/srt/models/llava.py,sha256=xrkg8sht8tBOID7427IEZtHL-KKWfEivDe2NqGjTSAs,26373
 sglang/srt/models/llavavid.py,sha256=dYUkKfHoE15vF_VXA_s_ICCTUMSmSgvP181fk8dUi0g,12185
 sglang/srt/models/minicpm.py,sha256=ws4AqhOfAvYHGd04QuXCZel-Oxy9_vN4p4rTjs9RSz0,13723
 sglang/srt/models/minicpm3.py,sha256=YIKJDTpwjmpLlv1sNT93k2yZMvGQlI_H87czjf6QYyo,24707
@@ -148,7 +150,7 @@ sglang/srt/models/olmo2.py,sha256=aC7svioN7XT5owRxPrvhvWBNMON9QXGQBWJ1KHMyXeA,13
 sglang/srt/models/olmoe.py,sha256=Rw-3YrHWd90MZQFnmcfUQ-3wAaI0PCFKb0DIrCDND3s,15347
 sglang/srt/models/phi3_small.py,sha256=44_my3QmgJ2N7SOkGZzEb62DXBeCVHojfmCWgkk2uCI,14802
 sglang/srt/models/qwen.py,sha256=_FKDbwaS5C07uJyyivZpBrXJVej4Ph9ivzJdzWJPxJ4,9904
-sglang/srt/models/qwen2.py,sha256=Kh6mW0H2jQdrPS9dJnJShLpo0BNEq6oI4oy5VMHGzac,12444
+sglang/srt/models/qwen2.py,sha256=be4xgcuqNa9kBdaL7x3PjsnUky6fh5K33c_khAWSi04,12959
 sglang/srt/models/qwen2_moe.py,sha256=rYUk_vZW3ftKIIlqPvJZ1K-6oZ_PfGspixh1zm2Y8C8,16538
 sglang/srt/models/qwen2_vl.py,sha256=3EaUlTbyWOTRXA7eViK1WqmVbCFhXLIpnos49zzf-yM,26561
 sglang/srt/models/registry.py,sha256=inKh9iwOp3LFYm3nqujg-OtABClOP-ifc1stA9cZegA,3434
@@ -157,8 +159,8 @@ sglang/srt/models/torch_native_llama.py,sha256=YeXHorFm6QfnczLXwPb5TG9a-He0uiA9R
 sglang/srt/models/xverse.py,sha256=Oq--KqvbYu2H4TMVGEHpSnJLEwXBpxlncR9ilsQeckc,13579
 sglang/srt/models/xverse_moe.py,sha256=AawKEQw--oAl-yzwCjoaZRG7q3rdkyDiam3FS0zjf_c,15537
 sglang/srt/models/yivl.py,sha256=88OubtuZ38Dxb2LzfV_MTPBI4wKhh4NJqFu--efbhFM,4809
-sglang/srt/openai_api/adapter.py,sha256=gZEaG1dVSFv9WLj0369Ke1yrNNgi_gpgKxPt5Ju9mUw,53775
-sglang/srt/openai_api/protocol.py,sha256=4T9hGCrpfCUSBjKZFvemTfj49CkTUzpCcx6izLv3ir0,10246
+sglang/srt/openai_api/adapter.py,sha256=dvKq4O3Rhd77ad6iCtPNykgnk9PVJE-E8wHVsBAfCQQ,53927
+sglang/srt/openai_api/protocol.py,sha256=ecRNNqkhwwKZaIoJlPhtp2VTcHxBJDbNN8lrKS7uBx8,10406
 sglang/srt/sampling/sampling_batch_info.py,sha256=s--zNjk-LErZ5lMqnZ7KiuJltaziKRbQAU5qYpKIxAc,8564
 sglang/srt/sampling/sampling_params.py,sha256=n7RbBg_bS5fYhsiWa8uJYnfoXy_i5DvtTBOkuFnHDNU,5286
 sglang/srt/sampling/penaltylib/__init__.py,sha256=5vQw0Y5DSzmsoFg1IdMIKLwFVhYZ5ArADHVBYbSmOec,513
@@ -180,10 +182,10 @@ sglang/test/simple_eval_mmlu.py,sha256=FkwamjGMjueTixymkedF-YiPloSLiy4ftILFUrKZ9
 sglang/test/test_activation.py,sha256=jkdNRzJnbd5OgZliQaIXpxovlcky17UrweomcOcMxoE,1442
 sglang/test/test_layernorm.py,sha256=IacByD5d-stXjzBz8Ypamc7povlcedpKPbb_4JLgo3c,3720
 sglang/test/test_programs.py,sha256=1Z0umrsUu9pagzyGH5SrXl_qhKSyTfUv_kWC2mcn0qo,18208
-sglang/test/test_utils.py,sha256=0lY3ZNfS3JCB4LqSRJgBfB8I0MA8TUT-BJmnrvQC8vw,23797
+sglang/test/test_utils.py,sha256=HJG7kUQOk6n9FBbH89PDtQ41C3kt1cfJODhAEcFT0AQ,23823
 sglang/test/srt/sampling/penaltylib/utils.py,sha256=CjxHgywh0hx_87iynzQt_ztHu6zBVuE-YrZ-XPmW6U4,12906
-sglang-0.4.0.post1.dist-info/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
-sglang-0.4.0.post1.dist-info/METADATA,sha256=Vfw4EHmIeHM0mE5Eicf11T1xPVOMCpNXsGz1LjsoPO8,22427
-sglang-0.4.0.post1.dist-info/WHEEL,sha256=PZUExdf71Ui_so67QXpySuHtCi3-J3wvF4ORK6k_S8U,91
-sglang-0.4.0.post1.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
-sglang-0.4.0.post1.dist-info/RECORD,,
+sglang-0.4.0.post2.dist-info/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
+sglang-0.4.0.post2.dist-info/METADATA,sha256=maHXecD3U1DdhzfU2aBMhN96MQRqCBPsIA1KlO7t7dg,22512
+sglang-0.4.0.post2.dist-info/WHEEL,sha256=PZUExdf71Ui_so67QXpySuHtCi3-J3wvF4ORK6k_S8U,91
+sglang-0.4.0.post2.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
+sglang-0.4.0.post2.dist-info/RECORD,,