sglang 0.4.10.post2__py3-none-any.whl → 0.5.0rc0__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (105) hide show
  1. sglang/bench_one_batch.py +113 -17
  2. sglang/srt/configs/model_config.py +35 -0
  3. sglang/srt/conversation.py +9 -5
  4. sglang/srt/disaggregation/base/conn.py +5 -2
  5. sglang/srt/disaggregation/decode.py +6 -1
  6. sglang/srt/disaggregation/decode_schedule_batch_mixin.py +3 -0
  7. sglang/srt/disaggregation/mooncake/conn.py +243 -135
  8. sglang/srt/disaggregation/prefill.py +2 -0
  9. sglang/srt/distributed/parallel_state.py +11 -9
  10. sglang/srt/entrypoints/context.py +244 -0
  11. sglang/srt/entrypoints/engine.py +4 -3
  12. sglang/srt/entrypoints/harmony_utils.py +370 -0
  13. sglang/srt/entrypoints/http_server.py +71 -0
  14. sglang/srt/entrypoints/openai/protocol.py +227 -1
  15. sglang/srt/entrypoints/openai/serving_chat.py +278 -42
  16. sglang/srt/entrypoints/openai/serving_responses.py +1273 -0
  17. sglang/srt/entrypoints/openai/tool_server.py +174 -0
  18. sglang/srt/entrypoints/tool.py +87 -0
  19. sglang/srt/eplb/expert_location.py +5 -1
  20. sglang/srt/function_call/harmony_tool_parser.py +130 -0
  21. sglang/srt/hf_transformers_utils.py +30 -3
  22. sglang/srt/jinja_template_utils.py +8 -1
  23. sglang/srt/layers/attention/aiter_backend.py +5 -8
  24. sglang/srt/layers/attention/dual_chunk_flashattention_backend.py +1700 -0
  25. sglang/srt/layers/attention/triton_backend.py +85 -14
  26. sglang/srt/layers/attention/triton_ops/decode_attention.py +17 -0
  27. sglang/srt/layers/attention/triton_ops/extend_attention.py +143 -98
  28. sglang/srt/layers/attention/trtllm_mha_backend.py +332 -0
  29. sglang/srt/layers/attention/vision.py +13 -5
  30. sglang/srt/layers/communicator.py +21 -4
  31. sglang/srt/layers/dp_attention.py +12 -0
  32. sglang/srt/layers/linear.py +2 -7
  33. sglang/srt/layers/moe/cutlass_moe.py +20 -6
  34. sglang/srt/layers/moe/ep_moe/layer.py +77 -73
  35. sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +101 -12
  36. sglang/srt/layers/moe/fused_moe_triton/layer.py +416 -35
  37. sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py +188 -3
  38. sglang/srt/layers/moe/topk.py +12 -3
  39. sglang/srt/layers/moe/utils.py +16 -0
  40. sglang/srt/layers/quantization/__init__.py +22 -0
  41. sglang/srt/layers/quantization/fp4.py +557 -0
  42. sglang/srt/layers/quantization/fp8.py +3 -6
  43. sglang/srt/layers/quantization/fp8_utils.py +29 -0
  44. sglang/srt/layers/quantization/modelopt_quant.py +259 -64
  45. sglang/srt/layers/quantization/mxfp4.py +651 -0
  46. sglang/srt/layers/quantization/mxfp4_tensor.py +133 -0
  47. sglang/srt/layers/quantization/quark/__init__.py +0 -0
  48. sglang/srt/layers/quantization/quark/schemes/__init__.py +6 -0
  49. sglang/srt/layers/quantization/quark/schemes/quark_scheme.py +55 -0
  50. sglang/srt/layers/quantization/quark/schemes/quark_w4a4_mxfp4.py +118 -0
  51. sglang/srt/layers/quantization/quark/utils.py +107 -0
  52. sglang/srt/layers/quantization/unquant.py +60 -6
  53. sglang/srt/layers/quantization/w4afp8.py +1 -1
  54. sglang/srt/layers/rotary_embedding.py +225 -1
  55. sglang/srt/layers/utils.py +9 -0
  56. sglang/srt/layers/vocab_parallel_embedding.py +8 -3
  57. sglang/srt/lora/lora_manager.py +70 -14
  58. sglang/srt/lora/lora_registry.py +3 -2
  59. sglang/srt/lora/mem_pool.py +43 -5
  60. sglang/srt/managers/cache_controller.py +55 -30
  61. sglang/srt/managers/detokenizer_manager.py +1 -1
  62. sglang/srt/managers/io_struct.py +15 -3
  63. sglang/srt/managers/mm_utils.py +5 -11
  64. sglang/srt/managers/schedule_batch.py +28 -7
  65. sglang/srt/managers/scheduler.py +26 -12
  66. sglang/srt/managers/scheduler_output_processor_mixin.py +1 -2
  67. sglang/srt/managers/scheduler_recv_skipper.py +37 -0
  68. sglang/srt/managers/scheduler_update_weights_mixin.py +6 -0
  69. sglang/srt/managers/template_manager.py +35 -1
  70. sglang/srt/managers/tokenizer_manager.py +24 -6
  71. sglang/srt/managers/tp_worker.py +3 -0
  72. sglang/srt/managers/tp_worker_overlap_thread.py +3 -0
  73. sglang/srt/mem_cache/hiradix_cache.py +53 -5
  74. sglang/srt/mem_cache/memory_pool_host.py +1 -1
  75. sglang/srt/mem_cache/multimodal_cache.py +33 -13
  76. sglang/srt/mem_cache/storage/hf3fs/client_hf3fs.py +2 -2
  77. sglang/srt/model_executor/cuda_graph_runner.py +7 -6
  78. sglang/srt/model_executor/forward_batch_info.py +35 -14
  79. sglang/srt/model_executor/model_runner.py +19 -2
  80. sglang/srt/model_loader/weight_utils.py +10 -0
  81. sglang/srt/models/bailing_moe.py +425 -0
  82. sglang/srt/models/deepseek_v2.py +72 -33
  83. sglang/srt/models/ernie4.py +426 -0
  84. sglang/srt/models/ernie4_eagle.py +203 -0
  85. sglang/srt/models/gemma3n_mm.py +39 -0
  86. sglang/srt/models/glm4_moe.py +24 -12
  87. sglang/srt/models/gpt_oss.py +1134 -0
  88. sglang/srt/models/qwen2.py +6 -0
  89. sglang/srt/models/qwen2_moe.py +6 -0
  90. sglang/srt/models/qwen3_moe.py +32 -6
  91. sglang/srt/models/step3_vl.py +9 -0
  92. sglang/srt/models/transformers.py +2 -5
  93. sglang/srt/multimodal/processors/step3_vl.py +3 -1
  94. sglang/srt/reasoning_parser.py +18 -39
  95. sglang/srt/server_args.py +142 -7
  96. sglang/srt/two_batch_overlap.py +157 -5
  97. sglang/srt/utils.py +38 -2
  98. sglang/test/runners.py +2 -2
  99. sglang/test/test_utils.py +1 -1
  100. sglang/version.py +1 -1
  101. {sglang-0.4.10.post2.dist-info → sglang-0.5.0rc0.dist-info}/METADATA +16 -14
  102. {sglang-0.4.10.post2.dist-info → sglang-0.5.0rc0.dist-info}/RECORD +105 -84
  103. {sglang-0.4.10.post2.dist-info → sglang-0.5.0rc0.dist-info}/WHEEL +0 -0
  104. {sglang-0.4.10.post2.dist-info → sglang-0.5.0rc0.dist-info}/licenses/LICENSE +0 -0
  105. {sglang-0.4.10.post2.dist-info → sglang-0.5.0rc0.dist-info}/top_level.txt +0 -0
@@ -1,7 +1,7 @@
1
1
  sglang/__init__.py,sha256=x1v8748WXS2BAptWAd5fR3YBEkngC7e_P_rkCNuNCGg,1702
2
2
  sglang/api.py,sha256=rcp3GeoyZhmJ0GDLPRkuZNcxd0TBJy_wfUDpcmQoqW8,7210
3
3
  sglang/bench_offline_throughput.py,sha256=Bge0KG4Uiv9CHsxDnqKdXJmG_TcxM4blz3CmvLmDRvM,14815
4
- sglang/bench_one_batch.py,sha256=kSqUxn8kEib6GkEdSDZgAaY6P1JuFL2BadbfVuXJjQs,19610
4
+ sglang/bench_one_batch.py,sha256=gfb4hSLbRupVOd6wfmSMgcJqjse17eQbkMn_Q2xRmGA,23086
5
5
  sglang/bench_one_batch_server.py,sha256=LS1BRmFwP67IpYlU5wmkWjqquiA5drvtWe-fwHnFSKI,14170
6
6
  sglang/bench_serving.py,sha256=nOjDnqOKDezDvtCvEhC1_FAoWVkir38bSaEIcSGGVZg,71748
7
7
  sglang/check_env.py,sha256=qDMIG2rCNBH1yKnxQmF-Bp10oiFMUKMgfZLHZYOmdSY,8412
@@ -10,7 +10,7 @@ sglang/global_config.py,sha256=ZMTux_PsGnvkyJ0kTFwhTdbnFwIjjpGDogut_9Lu4Vo,1732
10
10
  sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
11
11
  sglang/profiler.py,sha256=tEHzHerXC-ymk4OrkoUcMbgcGHmb8VESthsNSP2Yx9w,4417
12
12
  sglang/utils.py,sha256=C_r3OF-OA6DwUz6KYQF6xWUBApjHaqSPsV9JccafrWI,16403
13
- sglang/version.py,sha256=Ypmiam91zyRF3ijHnORC1GWsK_sC2iafbARSRi4vW8Y,29
13
+ sglang/version.py,sha256=Ko_9h2aNaKSGsWwzeD1Y47Z_-706MS9StmJBcxrT3V4,25
14
14
  sglang/eval/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
15
15
  sglang/eval/loogle_eval.py,sha256=-CC2s2kh5qUoDrHRkQVkC_jNvBgNojXbf456ny5s78s,4557
16
16
  sglang/lang/chat_template.py,sha256=ZNcuDvDZE9781tQqfKpptck_zJVK0aggq7IxrNHltao,21934
@@ -30,20 +30,20 @@ sglang/srt/_custom_ops.py,sha256=9GPNUyqnybgvUnzdQ8n5_NgluHmQNw0Gptos0iLPfrY,534
30
30
  sglang/srt/aio_rwlock.py,sha256=6LYtOdeTUY3hkfa1dmYkgsaF2ttrwIF3hUWz2AZ2fqw,2970
31
31
  sglang/srt/code_completion_parser.py,sha256=KFa95OU0TeVjJkOqIgS4xV3kaJ5dFWOmAAgISyc1oEc,3803
32
32
  sglang/srt/constants.py,sha256=0i-tEwG2BSYNDy96MxnGHV5HnBELkYcnsVGsE-R18o0,93
33
- sglang/srt/conversation.py,sha256=0lMeOqBiVhmIwNjt9TxwyuraNk6XGpYKCpq4CLB5ai4,38730
33
+ sglang/srt/conversation.py,sha256=pli5tbV3AUBciFBqE9l4tXST2Wgem7vsjNlnwqUDLas,38936
34
34
  sglang/srt/custom_op.py,sha256=7OPQAZe3PqDXk-xM0Wxl7pm9cazilNJfkNiYIWztWgI,3198
35
- sglang/srt/hf_transformers_utils.py,sha256=dSMTGS1Cac25-5Ygj2rUZba78VFnXM0acyMAW5aU8NY,13059
36
- sglang/srt/jinja_template_utils.py,sha256=LHqdxzuYfUbRY-QV8tYYPejulf1RrxNy--M3JFLKEZU,7255
35
+ sglang/srt/hf_transformers_utils.py,sha256=e--dYjSaaB0t2FhrvHlUrd-l2KmrLIRRtyGErDAGV_Q,13786
36
+ sglang/srt/jinja_template_utils.py,sha256=T5jdZmXUh3I6Iw-CYYqGdc4MJJH17rmf24pvKcv6ky8,7488
37
37
  sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
38
38
  sglang/srt/operations.py,sha256=ddQ8KO63L73OciaR8MZ9h2h83gKVY4-WuWgeEGowPJA,5346
39
39
  sglang/srt/operations_strategy.py,sha256=-w6GI2VgSGvORy3pUWRAJo6244ykSW1ZBV1Dx0fZlHA,7138
40
40
  sglang/srt/patch_torch.py,sha256=OUPCGQSQz3MVZB1zZ_Eq8lXiw0uIKJ_HWjqQolI8FsM,3088
41
41
  sglang/srt/poll_based_barrier.py,sha256=9vjn32ROlSueEk7EuZjWKUrrL7XVR9KJsSVliGTYB_4,937
42
- sglang/srt/reasoning_parser.py,sha256=pN7SbLjaiyT5tdrAZgOFWy3QrDKnZEHkBFeIKFXknZM,9288
43
- sglang/srt/server_args.py,sha256=3DwQscmXEUNALT294DdVuMhXqjfA6CXKUqOEq-vXGxY,90227
42
+ sglang/srt/reasoning_parser.py,sha256=6OvK4BZ_L_JsrtgXZgFMeBccc3aCAYKEkiYNv7nCWo4,8515
43
+ sglang/srt/server_args.py,sha256=RhWAIFZ69MIXbmE9IQQe4J8QxLz_7EP_NPi_Th5T9nc,96012
44
44
  sglang/srt/torch_memory_saver_adapter.py,sha256=K_eTx0UU84MHSTXI3iqYLdHV4IWtJMJ2FKdGFJR8v1E,2417
45
- sglang/srt/two_batch_overlap.py,sha256=X5DE7EcOdr1XQQLzrZwzrfahqYzdBSvBjhnkfIClpEk,28964
46
- sglang/srt/utils.py,sha256=ILvrvwyiV6iCOVEJ7W331q4jTGzysD8hMl8ao50EVFY,92832
45
+ sglang/srt/two_batch_overlap.py,sha256=-jICdc_ysKkHIYfCSdMSFoHnZ8uSeo3I4lT6rhcdJtA,34170
46
+ sglang/srt/utils.py,sha256=ggT1LuNwFDLQp2dCGWKjFFIE9JsJ5VPzYaxlm_Zqy8Q,93664
47
47
  sglang/srt/warmup.py,sha256=zldxhMlXpclRAJXmfBjJNUJd1eDizVdysibBvQyTVuA,1782
48
48
  sglang/srt/configs/__init__.py,sha256=gas3cKYVgZp7qF4m8FejX4EUUmdQd_w5WDTm7fo7yM0,743
49
49
  sglang/srt/configs/chatglm.py,sha256=j-b0YkdYUmQm2y1kNmMJtKeACxWKmBbvNNkDWbs6kbI,2907
@@ -56,7 +56,7 @@ sglang/srt/configs/janus_pro.py,sha256=yI2c_aWBIKiTcK6RZxK6hq8CIxyYRH_NN8uaSNKsr
56
56
  sglang/srt/configs/kimi_vl.py,sha256=4W7VQI3pr888ZsFA2SqCQo4mI0seXTOrGQ-x3oTvWew,1358
57
57
  sglang/srt/configs/kimi_vl_moonvit.py,sha256=hx2Rt4JSFbvy2HUTeLjBpge87m8M6ITAhqsgdNf_Jd4,1163
58
58
  sglang/srt/configs/load_config.py,sha256=qs-AxuplouBx2tsv9KGBOLZPbwzuVA4vbktbGP_cRp8,3309
59
- sglang/srt/configs/model_config.py,sha256=opeGqkpJcCjPLNLSa_Wf2PfQPB17leJKYO_VgYi17q8,28888
59
+ sglang/srt/configs/model_config.py,sha256=u6BS-f4QFOrcjwQMm1P9qqIhQ3Kr1pf2woPqY_YYb8s,30288
60
60
  sglang/srt/configs/step3_vl.py,sha256=_Otgnym57DVgB_kZ__8c1_Ys5gSalA_K0ZuVjcG51T0,4845
61
61
  sglang/srt/configs/update_config.py,sha256=D-myV9Ombwp6wTF0RX9qUO6m5rxFiUcyse32FQADG28,4787
62
62
  sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
@@ -79,31 +79,31 @@ sglang/srt/debug_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3
79
79
  sglang/srt/debug_utils/dump_comparator.py,sha256=8ZPwP0zojfRHvPiSKgpESYFcC3tR-sBjIej5SotOmpg,3947
80
80
  sglang/srt/debug_utils/dumper.py,sha256=eyrg5lmwyYF88kn8b06kpu1__aGnWb6CLM8fxywtooI,3158
81
81
  sglang/srt/debug_utils/text_comparator.py,sha256=pg58q3fzsnd3rJKRgZTdcOviOtAv8pSYZdwf-TLffMM,5522
82
- sglang/srt/disaggregation/decode.py,sha256=xPXKHz0e8O9GZetHiGVQoN8iIwEA1GwO8jj0NkCzgp4,33884
83
- sglang/srt/disaggregation/decode_schedule_batch_mixin.py,sha256=R3JmjHHU_j6aOtUkilHsL5hzNn5oJpZUqmX_mEsT0Ek,6485
82
+ sglang/srt/disaggregation/decode.py,sha256=w8CpZAN6EqNI4NncRkwEEUERTrRlDbghYUw85jnatMU,34126
83
+ sglang/srt/disaggregation/decode_schedule_batch_mixin.py,sha256=Iu7ehHGAfYIVATtClwX6y-6zxWs-JyDlMotYo_5yuUc,6598
84
84
  sglang/srt/disaggregation/kv_events.py,sha256=yFbtwOeblcCmOXTjg00TopxnyjkmCBQIVz46KB8jetY,13555
85
85
  sglang/srt/disaggregation/launch_lb.py,sha256=Elw-J-97jTA70Ml1EMp_J-k9X9H5WHTWOC2XEvmyJRY,3827
86
86
  sglang/srt/disaggregation/mini_lb.py,sha256=bpnz6d90nv7vAoiPxmxFKlM1Gjct0R4IPkF9SwwBk9I,14504
87
- sglang/srt/disaggregation/prefill.py,sha256=03li8dKrOFfhUsvaSuGKQWgH9eAyWaYv59cBiWpzo-s,24136
87
+ sglang/srt/disaggregation/prefill.py,sha256=dFK78uKPWpridSB7RCrMUJRV-j8Ef2fWZCX2RFoOArE,24231
88
88
  sglang/srt/disaggregation/utils.py,sha256=ql5p9yjBCRcR0YMgcgRK6tH0kPGvLNTUgeXBr_BUcoI,12198
89
89
  sglang/srt/disaggregation/ascend/__init__.py,sha256=-lxnica6fZYmYNpKwmE8yESUpQ5Sxf2DiZoBHzboKc8,146
90
90
  sglang/srt/disaggregation/ascend/conn.py,sha256=FOYrGtEhj58iTjhYT1eP96DpyxBLFFR87icqYCzWmCY,1215
91
91
  sglang/srt/disaggregation/ascend/transfer_engine.py,sha256=_4PszsQL6wICdg5cqpR2N9TbZpUY6bBX8tgE3xeaYfc,2137
92
92
  sglang/srt/disaggregation/base/__init__.py,sha256=4VwUv0aWxwmVL1049XK82aLTNxmt0WY5RPy9li-wyVk,160
93
- sglang/srt/disaggregation/base/conn.py,sha256=CPDAoAkYaFtVPLa1QROfwipSVe7MH6omzIBHzo8TSYk,2811
93
+ sglang/srt/disaggregation/base/conn.py,sha256=UlPo4M7m6Lk1VQBIN2kQ1WlGF9IoQFId5I1PN-era3A,2872
94
94
  sglang/srt/disaggregation/common/__init__.py,sha256=7yl-EGLMVKRpBUaGF_7lwAsw2J_mqpRZV0238VGxD9o,126
95
95
  sglang/srt/disaggregation/common/conn.py,sha256=3YJVviObY3fkSfQbVxE5lgcqMmqM5nDjWfEjS5-SW6I,17054
96
96
  sglang/srt/disaggregation/common/utils.py,sha256=SxRhAWisNK8seGhb5BXBJ5u53DF7yeKVPMWPcB5ywbE,1194
97
97
  sglang/srt/disaggregation/fake/__init__.py,sha256=jJGWdXwaQiGIoR6atKqkQfkJmVyQ09l55VUN2WjwaeY,77
98
98
  sglang/srt/disaggregation/fake/conn.py,sha256=oD1DArn1yDFZCu-X6p93uSLlAXEkt9lYxERICMznxGw,2286
99
99
  sglang/srt/disaggregation/mooncake/__init__.py,sha256=0TgqkAdQI1YynbHY6c0QISvVoOSk-0SwCIq5rjPSmgE,156
100
- sglang/srt/disaggregation/mooncake/conn.py,sha256=BrOcdnqk7Ut90qDqajNzQx3qTBlnsVhLmMgUgXVGCdg,62677
100
+ sglang/srt/disaggregation/mooncake/conn.py,sha256=5GarzDKjUgjEt5-e_hl8jufhwdtOSlcbzFbmrMEzeBY,67155
101
101
  sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=-ulLhz6DBDsmEjwNjjH5Pj8ngZKISgwBc_zL3ViGDN0,4761
102
102
  sglang/srt/disaggregation/nixl/__init__.py,sha256=qODVPIGWUXKXq4zsRIcMYoAoAeg6nBIN9vdQOlVMANE,136
103
103
  sglang/srt/disaggregation/nixl/conn.py,sha256=eSof87fG21Dd4COszfnbeXIxne3TWvw0mSvCOkjsBZc,20323
104
104
  sglang/srt/distributed/__init__.py,sha256=jFOcyt-wFAPMBUAf9zkZalNQlt-4rqmT6pCKBz1E4qo,149
105
105
  sglang/srt/distributed/communication_op.py,sha256=IBnFUdMftK_VSTMMMitGveonorFUUVNL4guqO31cMSc,1130
106
- sglang/srt/distributed/parallel_state.py,sha256=NAIEJWT5IgMqz-Kpb0SKgok2HElfXWWTOABUqx-t53Q,61675
106
+ sglang/srt/distributed/parallel_state.py,sha256=90eqCuQYgA1OI611KGznBAXVVtTb3XYdeXus5WrvohI,61659
107
107
  sglang/srt/distributed/utils.py,sha256=aaCxATncLGnVgB0WlGpBdee0behKW8Dy_dakqcuKSaQ,8497
108
108
  sglang/srt/distributed/device_communicators/cuda_wrapper.py,sha256=3jvPG-Ow5UBLiXhfx8T8snR7crSZbPpARAggsDPWq7k,7038
109
109
  sglang/srt/distributed/device_communicators/custom_all_reduce.py,sha256=p-GLYYnMSjcUvK30qcwYhXEJwGrOz6rR99p_23SWFj8,16570
@@ -118,23 +118,28 @@ sglang/srt/distributed/device_communicators/quick_all_reduce.py,sha256=4j1_E4azo
118
118
  sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=IrSrnpZnii0EJTS1CYRwEwE7gyHxJBVgI2QuJS3AKW0,20906
119
119
  sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
120
120
  sglang/srt/entrypoints/EngineBase.py,sha256=yKN76witT2jz1zhmLHmPNLGMpK2UiOTaKQ2KPD8l99U,2594
121
- sglang/srt/entrypoints/engine.py,sha256=P0K6pCyznxDheUVmzLhmQPTX4arVPQY0PAGQKQhwZNI,31856
122
- sglang/srt/entrypoints/http_server.py,sha256=qdi-Qf3UQnufdRMx4m-MaBsN1maJ3oN3VQldiNFyOec,39069
121
+ sglang/srt/entrypoints/context.py,sha256=104mtyJw9ZAplRpliLPhTWbvYJ7NVRm8xlegMZuFP2A,8006
122
+ sglang/srt/entrypoints/engine.py,sha256=lfix_7Uyz9-SYL4bqr_r1OxpHT-BUHFP-vBkXEr7d2E,31903
123
+ sglang/srt/entrypoints/harmony_utils.py,sha256=vuI-KsX99Czj9YhkEWPMr1TcMz7hY57IFXeW4y1Jt8o,13403
124
+ sglang/srt/entrypoints/http_server.py,sha256=H-MGK3RH6t-90x4pyRuvaaCh6HpnRL-CSdYHwtX4mqo,41426
123
125
  sglang/srt/entrypoints/http_server_engine.py,sha256=_--j4U04OeJLlnnv1f0XmCd_Ry0z1FlhkrbePX8rYV0,4938
126
+ sglang/srt/entrypoints/tool.py,sha256=6fyXArrLqDsZO3rTtPfZ_6ngYLfTMRyyTSvESqos5tA,2678
124
127
  sglang/srt/entrypoints/openai/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
125
- sglang/srt/entrypoints/openai/protocol.py,sha256=2XfURpoICsk4UZs4W9lm2A1yXNC-iwqSdZOuof0FXjo,19183
128
+ sglang/srt/entrypoints/openai/protocol.py,sha256=nMbnURu9GN8OcvmLDlXFwHxGul4JlNKHj8bPL5srmI8,26748
126
129
  sglang/srt/entrypoints/openai/serving_base.py,sha256=epD6J7K72p6TP8o2dGYf-a76nXeF80vFhPRpWts6684,5445
127
- sglang/srt/entrypoints/openai/serving_chat.py,sha256=wscn5l3Kcgk0fqhxke31_ALc8h3mptwdIzYX5dpTWEI,38853
130
+ sglang/srt/entrypoints/openai/serving_chat.py,sha256=zV_a2r_hmP576_z5QzSvNEBnTgOY5xzlbBiB72LpaQU,48675
128
131
  sglang/srt/entrypoints/openai/serving_completions.py,sha256=emIutVmnJgOgGRuAqdB80-kVHUfQbE67n1VS-76o4QY,16690
129
132
  sglang/srt/entrypoints/openai/serving_embedding.py,sha256=hqPt2ELpT1yQ8sBXo801aLJ3sExoPXs_K5ZQtC2vpAs,6274
130
133
  sglang/srt/entrypoints/openai/serving_rerank.py,sha256=8n6y5kC6UhRpWrDtzH0eIp0dznW12FT60MLCR4bvD2U,3333
134
+ sglang/srt/entrypoints/openai/serving_responses.py,sha256=65KIJ6gjEUz_kbmqejMVY45NixVTz4VqL2zZIMilk3Y,53394
131
135
  sglang/srt/entrypoints/openai/serving_score.py,sha256=ebAts-m6Pq-LTgFKwggkywBUrAgUSppHHVEzgwYHUzo,1955
136
+ sglang/srt/entrypoints/openai/tool_server.py,sha256=qbhkCu8CkDv1kLUb-2778Mhb2HfrAKvzserYZHxxWag,5648
132
137
  sglang/srt/entrypoints/openai/usage_processor.py,sha256=9LTB5rqdRuMKyZrIXiUBuF_WKaSg9X45YdzERDxbtCY,2746
133
138
  sglang/srt/entrypoints/openai/utils.py,sha256=DT9quW0F1DX1P9NAuMxpERgXcNYKKtm7rd7J1m7EaDQ,2112
134
139
  sglang/srt/eplb/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
135
140
  sglang/srt/eplb/eplb_manager.py,sha256=Wt7w4L5g1BSsXx1mqoGo8WjkB55f4nGB4KqC3y_ijl8,3394
136
141
  sglang/srt/eplb/expert_distribution.py,sha256=PG40hrCP2-RJR6hmS80I0N5HUj2XZnxR0CBBFxq9xvg,31740
137
- sglang/srt/eplb/expert_location.py,sha256=A9zkr4lCl8480WlPVIMosdjSrMI1y_ym_v1iRIjJP68,16753
142
+ sglang/srt/eplb/expert_location.py,sha256=SIinJLiDhS4mctTiMdIQzQZFchwrrvES0ufEVJ4OGeg,17076
138
143
  sglang/srt/eplb/expert_location_dispatch.py,sha256=vA0o-7bXf847BQAVCqElWKEpofZLAIsir_d0pew2KNQ,4178
139
144
  sglang/srt/eplb/expert_location_updater.py,sha256=eMtJVMCRklfvKPbFiDKy85f5J9s_krJbpRCSArYZjVs,21157
140
145
  sglang/srt/eplb/eplb_algorithms/__init__.py,sha256=c6fo4qM8N0p5DyyfGp9fxqc2wkxvzSJBzJFRG1I1leg,1959
@@ -148,6 +153,7 @@ sglang/srt/function_call/deepseekv3_detector.py,sha256=WzJbwCe8DoOi8guX9pdVMd70C
148
153
  sglang/srt/function_call/ebnf_composer.py,sha256=HvLTgfNnBNKFKW-L36DKh07yRLbdTYjSfbTcLRUxZUc,14948
149
154
  sglang/srt/function_call/function_call_parser.py,sha256=t4ISHoB_xF7HFy7DF8zCJ0U8EwWivbc_DNlLX1o8Xb0,8107
150
155
  sglang/srt/function_call/glm4_moe_detector.py,sha256=TD67qS7T-9iR4-P5fhrUbqqYWEYVTb30dDNd7lGj4Rk,6934
156
+ sglang/srt/function_call/harmony_tool_parser.py,sha256=dGgjXYEULlnxfORsJdiv0OvqW9O15Jy_deFwP7-wiMc,4651
151
157
  sglang/srt/function_call/kimik2_detector.py,sha256=7unW_GwH7I6jrh2BZbw22mWRZzQ0OMTYDz_xVX5abPU,10116
152
158
  sglang/srt/function_call/llama32_detector.py,sha256=trZja7IXc2IWUko5PRj7p4pRU6XCVjtxr-7qJNr0EWo,3647
153
159
  sglang/srt/function_call/mistral_detector.py,sha256=Ts2HrJdu5r2lCCMVKPZfnCds9pt1K5H1EaOQrF_HD60,4686
@@ -158,27 +164,28 @@ sglang/srt/function_call/step3_detector.py,sha256=SAibrispu0peDQzqOG2-V93KGHtV6U
158
164
  sglang/srt/function_call/utils.py,sha256=__ImDF2kNyoLWsYO5RYoryvy1mmgEjnjXlCvLv-uLCM,1695
159
165
  sglang/srt/layers/activation.py,sha256=uhfhh10N5iLoLEBCvFPRjCfDxu5jXe1ehHJb3ISIjig,7840
160
166
  sglang/srt/layers/amx_utils.py,sha256=1mENgHK2B8mgaD1oMtgbZ15Jmy_Uu1QueBmo09Ff2iA,2865
161
- sglang/srt/layers/communicator.py,sha256=xzVeuW2ymLERcKgqVNLl8Hv7hBr0_scc-b4Vx1-wpOA,20453
162
- sglang/srt/layers/dp_attention.py,sha256=tOre7il5Cppu930-tzM6d4AgjHuIJMCMlg9hJkuBcjA,11382
167
+ sglang/srt/layers/communicator.py,sha256=beXuwnzexWN0J82rmv8RrvJF9rc97nSeOu3vjnb5RpA,21336
168
+ sglang/srt/layers/dp_attention.py,sha256=vm20fQ60OfdrXxOylsLAEA_NQC3yekalQ3NFWctqCfw,11949
163
169
  sglang/srt/layers/elementwise.py,sha256=MyQUflyKEfPZ-BggW1Kd4hB53RFD6FXGc2S5LXjx_do,16026
164
170
  sglang/srt/layers/flashinfer_comm_fusion.py,sha256=fkTcAB7qYwSWi95qI3Rqq0JUyDpJdcYkd4TYkWO01HI,5891
165
171
  sglang/srt/layers/layernorm.py,sha256=Qkx4Jzb3KWOt2okAVM6Is4FDClpUIJShcPGoElUSnEQ,9598
166
- sglang/srt/layers/linear.py,sha256=Fp5xLw1kCkj-zpRc9_YH4B7TNkYcG4E0WURkzc-Fbwo,52706
172
+ sglang/srt/layers/linear.py,sha256=hQE5p4hIC2qAqlGM2jtkHGVGxaD9Drt17FFHuPZQ_K4,52461
167
173
  sglang/srt/layers/logits_processor.py,sha256=LToJR70CfI7_gBLVNLmn5kUTovj3BU_7UDc4hjlac08,27077
168
174
  sglang/srt/layers/multimodal.py,sha256=wVo5-OLz-J4k8fM0ld1ZKMtzOsSue4aa-uILak-I8Bk,2398
169
175
  sglang/srt/layers/parameter.py,sha256=jCg0G-12GZqTa9gGOqEtVCsnnuBGnYg0VmF6pc6oCFg,17455
170
176
  sglang/srt/layers/pooler.py,sha256=uZ6WX1FLMEafZwusyZdm6KuVlIwSjbKrdwk2qzgqNGk,3812
171
177
  sglang/srt/layers/radix_attention.py,sha256=vtzCPfEiWYunglcLfqO9dcATb-LXzU4sFBwIWsWHbiY,3513
172
- sglang/srt/layers/rotary_embedding.py,sha256=EhxI0E8jcTWZ2COpnku7crbW8Hew5fe_ujMndj7hKvE,52246
178
+ sglang/srt/layers/rotary_embedding.py,sha256=IoWhqKpBVylRdmjCDZQ8_5fTVPp6UzqnCSgJrWNhQRI,60992
173
179
  sglang/srt/layers/sampler.py,sha256=xNds1migup2s6b9_pS6ljkJUkvNtv7nmTGeIdOzoQ6w,11182
174
180
  sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSahyMfA,4063
175
- sglang/srt/layers/utils.py,sha256=IWGg1Hb7c33Z3LHRPVJyUAzp3BnSid23ZWXAmJ_Jvp8,1204
176
- sglang/srt/layers/vocab_parallel_embedding.py,sha256=UT0GFg94Kwvs9RFRdKR9ISBsftcXycP7s_nZQnf_urs,22590
177
- sglang/srt/layers/attention/aiter_backend.py,sha256=7sEUgViw-xl3yok91yyOD9gTi8lQmME0g0ZiKVTCcyI,32851
181
+ sglang/srt/layers/utils.py,sha256=NEWsf7OlUgcX42WHXxANduDL0ByBm3beirCp8gvHDRs,1437
182
+ sglang/srt/layers/vocab_parallel_embedding.py,sha256=x0421kNUgnAb5LnkJR6O67vGZ0ikSD-ThluV3QGsZ_8,22673
183
+ sglang/srt/layers/attention/aiter_backend.py,sha256=2ThUq2lypJuQ9g06OxjwwqbZNplnQ_pkPkz8OVr1CVQ,32735
178
184
  sglang/srt/layers/attention/ascend_backend.py,sha256=jPCsU9_gH1iZNoZHD9nCeDdVdXqBt31LI65N55BTJPg,8250
179
185
  sglang/srt/layers/attention/base_attn_backend.py,sha256=_vM05ddWy6SaoqWYPR8w5lflSE4MhcVFR271Q9EKK8E,3600
180
186
  sglang/srt/layers/attention/cutlass_mla_backend.py,sha256=PB5BjbJHhmTg_t0kXfarTI6C3pPvKhBAqvCSGGajou0,9846
181
187
  sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
188
+ sglang/srt/layers/attention/dual_chunk_flashattention_backend.py,sha256=9Jyb3XhVwZLz7I6B8sjHi0pCwypBn0rqYxe_CKVEYXY,68635
182
189
  sglang/srt/layers/attention/flashattention_backend.py,sha256=22vHnwDBDTtx0WoW76TsYUPewPu9OqCYQtJmLKqVJ2s,93917
183
190
  sglang/srt/layers/attention/flashinfer_backend.py,sha256=HQp14R29hQVXYjATKPoGaHY9LJ9e1cw2k-RhBEmFxt8,51505
184
191
  sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=NirRlNFxD0S2EncvvmrerIxNfuTnYpvemZV3MR5_FnU,34040
@@ -188,30 +195,31 @@ sglang/srt/layers/attention/intel_amx_backend.py,sha256=hH9lxXs_o66_IVs58oPhRCOX
188
195
  sglang/srt/layers/attention/merge_state.py,sha256=OnWmNV6LYTUj0y8SCy_IK7A9q6v9bZwR4_XiLOIvbrk,1423
189
196
  sglang/srt/layers/attention/tbo_backend.py,sha256=CBnSDUmDvT-Tbhi9NHNo1LLkgUAcIsWMJgKa-sAIKW8,9318
190
197
  sglang/srt/layers/attention/torch_native_backend.py,sha256=RsI4lYiQJAj-Xy0vKkzbyKOWHAVPOD9b0n0dEehwlos,9496
191
- sglang/srt/layers/attention/triton_backend.py,sha256=AryrPc4kzioxkiucPkApMA__0HoQTKOHNpwN0GEgm98,36923
198
+ sglang/srt/layers/attention/triton_backend.py,sha256=eBWkuNiHtB4bvm_fjKCqQ53mF3RAgaLFoRJbmSyK2qQ,40159
199
+ sglang/srt/layers/attention/trtllm_mha_backend.py,sha256=Q1DkrnUAC4ZX42xa_gK7nfc_sTE_4k-LOiyIGVWb-o4,12679
192
200
  sglang/srt/layers/attention/trtllm_mla_backend.py,sha256=KwKvpnL6CpO84oDRuLVnGfNtT3SQ1SRYCZRI0ElGabA,13446
193
201
  sglang/srt/layers/attention/utils.py,sha256=lx_jNw2VwlBe2VnDb1eXRRdN9WArw9rdWHTHqndUEFk,3174
194
- sglang/srt/layers/attention/vision.py,sha256=8eRhnwpUW646j03pAhyswjpmHSpDWryv7foIoDI0cNs,18828
195
- sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=eN0Lqpu5h27lMPuRnisX4jrU0WR061pCFSrh3vLlSD0,18936
202
+ sglang/srt/layers/attention/vision.py,sha256=GEmwwsh08WSuhm5V79FO7i1aSQefM8DgPnsTanwUZxM,19126
203
+ sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=pS7ZmwMoK5Hrl0d9-V7ZcGZ0YkOeQUAwwZLmONqBssE,19305
196
204
  sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=FbgrPO5GvEGEvojV7Ss0jX7384e9vuzX2goDCQuP1Zk,31124
197
- sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=ImpCuxszoeVdr6wzaLR3OGW3-50P9EAEvsRCdUVIrxE,14261
205
+ sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=2t80inXyHE2A4TWKUHP1SvYZzmM7cklRk9uCygONxXg,16010
198
206
  sglang/srt/layers/attention/triton_ops/merge_state.py,sha256=v9nD01a5eTnkwZxMwERtrrRfC5rs6GxkYOpErkAPcYI,2877
199
207
  sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=waZsmpKIp8rTgFSoM4QMabJuLaB3yW6ltOzAKJksBoE,6260
200
208
  sglang/srt/layers/attention/triton_ops/rocm_mla_decode_rope.py,sha256=664WnAJ91EiCUZOcnVDfbTQf4uGJ4ZDZB1CbxpEUFZc,13866
201
- sglang/srt/layers/moe/cutlass_moe.py,sha256=jprUaoU1PbfwiEP_yTE5Z8ulviQoZWI8tRk0lsNABzE,14348
209
+ sglang/srt/layers/moe/cutlass_moe.py,sha256=CtuoUWxgSfpFeSnJU4NepUVahGxEAp7-TFAbImAhJws,14987
202
210
  sglang/srt/layers/moe/cutlass_moe_params.py,sha256=9NRCmgP_Ug3gGqCcpi-x-QRbLjCNpw8792gKXwZsbEU,6522
203
211
  sglang/srt/layers/moe/cutlass_w4a8_moe.py,sha256=bc8s5Oc36pCF0VozxZuqFOOU2fov-4wGPlyduye3Imk,7296
204
212
  sglang/srt/layers/moe/fused_moe_native.py,sha256=SkOFgU23aZfHmLhnkVhdUWGx0-2wbo6uu9ohKBp2yOI,3486
205
213
  sglang/srt/layers/moe/router.py,sha256=UrPieRvemN7Ew48gtG7DA2xhNDBRSnZxzugTEBI-0_E,12006
206
- sglang/srt/layers/moe/topk.py,sha256=ttPFvXzTY0eyh6UH8fJZCjyDp-R9EssSP8m229MxQ_4,26925
207
- sglang/srt/layers/moe/utils.py,sha256=UcHdqZi57kgDRjF6nWnw7lZKGj877IVB4VaSK8qNtC4,1051
214
+ sglang/srt/layers/moe/topk.py,sha256=5O0Q5FlsjEyqo41jjDf7noFjhnyUZE5Ysw0TiyLapu4,27235
215
+ sglang/srt/layers/moe/utils.py,sha256=XOl_KRa7G3YOiDI4pP8MWPC_fDubxk-FUANIkK2Jm2g,1545
208
216
  sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
209
217
  sglang/srt/layers/moe/ep_moe/kernels.py,sha256=UFH-XacP3aBjAcn6_SeC641jGzjSrY0fz50IB9URix4,42706
210
- sglang/srt/layers/moe/ep_moe/layer.py,sha256=M4X5HOu5k0mumtzG2cif7OLuM08IKk8nyCDED08qKjA,26261
218
+ sglang/srt/layers/moe/ep_moe/layer.py,sha256=BfVqJ5vTN0FhJbq49JGtQQEoBVdsAiU5AmuvfFv1V_w,25830
211
219
  sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=4NXZHbCw-G-uSnNUj4up0yh3xBDPnT-x0pdoIr0lku8,831
212
- sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=Yb7LvdCQWonJC05_4UOMt3jQYWmRl-AkdY7H5EVwKhM,57575
213
- sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=z7HD4oNe7fITnlmmFMS4L6gf4GcpS_9CLXJ-B5dAh7w,27874
214
- sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=q8o4puG3ous6wTMbQFzydQ3P5emZbsNKepNaveLSpL8,4956
220
+ sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=KEzRCWicAMBLzQN2IASzTcTCGD4VdAwanYiZ4ehxj6M,60613
221
+ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=F6fJIjsFbaRb9d_YapwbXBPkA-yBaVXo8YB3sd450gQ,42067
222
+ sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=fC9PGO2Z3XAQ8cVBcY0TSH5BvVB7UI319Vp4Lfx37ns,10814
215
223
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=iNGsE2ZeVnQEnN4A8UJ9Jv0d3hbRF2MJ9oBgjup5Szk,2737
216
224
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=JJN0hryyLr5Zv3dSS7C8cPFhAwTT6XxUVnBGMZvV6JA,2752
217
225
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=ouRyZ5PEMPP2njPftCNhs-1g1y6wueWLmhI7G1SjV1k,4131
@@ -388,28 +396,31 @@ sglang/srt/layers/moe/token_dispatcher/__init__.py,sha256=QUjIW1xImUsIf9M8Ze6ger
388
396
  sglang/srt/layers/moe/token_dispatcher/base_dispatcher.py,sha256=ibQnHlcnemt8mwOMj01qz360XaherXvTY0T6wGOeBQU,1273
389
397
  sglang/srt/layers/moe/token_dispatcher/deepep.py,sha256=CbPI6Ul1GjrT81NMbhL2E_CSt6cwFFepMtmcEyTAv54,22830
390
398
  sglang/srt/layers/moe/token_dispatcher/standard.py,sha256=jseBfZpst5iCDxXxQV_-53qb0VMqy8RHqXJMf86Bdgg,430
391
- sglang/srt/layers/quantization/__init__.py,sha256=lMjOmhMntbY55RR5eKKLB338tdzlxGuBx-pT4f9jmIM,8130
399
+ sglang/srt/layers/quantization/__init__.py,sha256=wB1hLMUrdqEVAlkzKjo139rlRnlTt93_uoEYdZhNI9k,8692
392
400
  sglang/srt/layers/quantization/awq.py,sha256=gLEOx6QoGQ0McBEURXPVidKGsdNUjqc9sJwGLQx86rE,26690
393
401
  sglang/srt/layers/quantization/awq_triton.py,sha256=4XwJbRNlFmlvjbMGtWF1Wvk26Kf96162vNEmuI5Kars,11737
394
402
  sglang/srt/layers/quantization/base_config.py,sha256=CiSxG5iuOa-12V9kIOh0Fu8Y7knZxfFbbNGTo6nEaRE,7173
395
403
  sglang/srt/layers/quantization/blockwise_int8.py,sha256=3avetVUBjCiuYWBJQbElbVXWpx0SgsFsDc5F9F_zHgA,13879
396
- sglang/srt/layers/quantization/fp8.py,sha256=zo7Uh5De-jk3Wowdha52oWl70uc9sfbJgdNSTRrwDP4,48815
404
+ sglang/srt/layers/quantization/fp4.py,sha256=rmitvYd4PqDK906oeRgpaaKs_XIdDLMJ1UzSvG3nd-c,19317
405
+ sglang/srt/layers/quantization/fp8.py,sha256=Ee0c8ft9j5uUAhRM84Ng4oRE1hxfkK-YLBN_Fad0fvk,48753
397
406
  sglang/srt/layers/quantization/fp8_kernel.py,sha256=0iC_rlmWVEfZOtr5osgMrd8-QsW2pCi-paaX1NqoVdU,40694
398
- sglang/srt/layers/quantization/fp8_utils.py,sha256=6FVR6_Y8akP4ihCaGG_bLYurcC_PShPtiPgiEvGN5bE,25605
407
+ sglang/srt/layers/quantization/fp8_utils.py,sha256=SmJEVKQOqbR17vgfUbQqixvWbDGvD1lNBXZUIvvvj3E,26402
399
408
  sglang/srt/layers/quantization/gptq.py,sha256=3q9xfq5sJJHcrVr9C8-l9JdcLFvwKAAfWJEfZW6CzLs,39145
400
409
  sglang/srt/layers/quantization/int8_kernel.py,sha256=GfRC9FOn9exNvK4QHbUeBj3Hhv32VcyGphapFPt5b84,12625
401
410
  sglang/srt/layers/quantization/int8_utils.py,sha256=YK9CS-lb_n91kNCTKK5o5apYF31V2giDg5G5VKrpcUA,2356
402
411
  sglang/srt/layers/quantization/kv_cache.py,sha256=_9pF5rwvB7ta6Gdc5YKVVGbNzYwqmhIx4TrX1-xnodQ,3261
403
412
  sglang/srt/layers/quantization/marlin_utils.py,sha256=ShsoxcVTYo1Es3XKp9mlQNw-TsRF_o4mHIO5MfAP5HA,25906
404
- sglang/srt/layers/quantization/modelopt_quant.py,sha256=oj6jJ66s66kMCufuJapUxGaAX8PjuJB2pdzZuUDZcrM,38357
413
+ sglang/srt/layers/quantization/modelopt_quant.py,sha256=suZ49ni5ODCHq2qrAio2F59a4TCBmk87hjBGNr5n0Y8,46525
405
414
  sglang/srt/layers/quantization/moe_wna16.py,sha256=NGcFjiMXwfOX10dkHAY3EgZHxox2k7I402Bvwj3jW1Q,19105
415
+ sglang/srt/layers/quantization/mxfp4.py,sha256=8Ze0E1MqJL640C4ugLo7HlS32xFzULsxjToASYFnICU,24360
416
+ sglang/srt/layers/quantization/mxfp4_tensor.py,sha256=I1hh_uZqIqkgFnNyjaP17cDMLMV5rs3aFytg91MPhzk,5357
406
417
  sglang/srt/layers/quantization/petit.py,sha256=74Jn0eICq23v_b4rEctDaYTINtbqkdM6IXycT-_FbbI,8954
407
418
  sglang/srt/layers/quantization/petit_utils.py,sha256=-gy4zMhqNoIA1R0n7-5C0efV54jHonCUgPDUUFjGsyM,3245
408
419
  sglang/srt/layers/quantization/qoq.py,sha256=jM96uJfKz3vo1B6SSIrQXNT3vnD3UJrepK2eY-tSQU4,8139
409
420
  sglang/srt/layers/quantization/scalar_type.py,sha256=nBk4THjeRWMGW3hKWRdrjfQ0gwpuBxmYqapDBaRpnr4,12395
410
- sglang/srt/layers/quantization/unquant.py,sha256=vgI2G1iHpMr2ys1-HaQ_YIXJN0z1m9uM6MpJh_1Ktrg,12108
421
+ sglang/srt/layers/quantization/unquant.py,sha256=zLShmKPEMW1OeuiwDzDteE7a2m7YlM7WVw0jWPSBgOQ,14481
411
422
  sglang/srt/layers/quantization/utils.py,sha256=9ZmfS0MMqueQ6zIxMIFoQy2OFBQ5vHjOHBgVvUn7ews,16213
412
- sglang/srt/layers/quantization/w4afp8.py,sha256=wa5t0g2IwVFcwHxZ213f5Vsnt8P4fum7WVoxvf0llpw,11092
423
+ sglang/srt/layers/quantization/w4afp8.py,sha256=NxAFT-pRnwMb_9v5UD1DdCNK2oqFRil9z32J2fcs_Aw,11092
413
424
  sglang/srt/layers/quantization/w8a8_fp8.py,sha256=I_W9A4Pnluaf-SaT8zq-tAJTt1acpxfE_kG5mqdc5cE,10250
414
425
  sglang/srt/layers/quantization/w8a8_int8.py,sha256=VXGAMKMlRU7k6c-WRXDExBTOa2oLODRFq-RIZXGsI-c,35013
415
426
  sglang/srt/layers/quantization/compressed_tensors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -576,12 +587,17 @@ sglang/srt/layers/quantization/deep_gemm_wrapper/__init__.py,sha256=esJMd0Yuj68t
576
587
  sglang/srt/layers/quantization/deep_gemm_wrapper/compile_utils.py,sha256=FaHKgY9y1oi3zlCmkorSK4r4dhPy7NiviVIhtabvzjM,11536
577
588
  sglang/srt/layers/quantization/deep_gemm_wrapper/configurer.py,sha256=zkMzBAqQTSwG9TlGPuwmHV1yN0v5Y_5Hgyo0CKQDVNU,741
578
589
  sglang/srt/layers/quantization/deep_gemm_wrapper/entrypoint.py,sha256=On_uvP1ThxgQFAzOWKoteM0vIOHqmtI127A4qbPtmbc,3264
590
+ sglang/srt/layers/quantization/quark/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
591
+ sglang/srt/layers/quantization/quark/utils.py,sha256=Aq85sXcnVN2IJnfwCe5t9hfkodcDNE9OmxoGwrmzzRE,3487
592
+ sglang/srt/layers/quantization/quark/schemes/__init__.py,sha256=z1LHBPqkxAS6Wp70DWwESOSixWpIhIQwz1IHcWDdURQ,167
593
+ sglang/srt/layers/quantization/quark/schemes/quark_scheme.py,sha256=b8-Djo-UQAwgxSWViKDoEP5dOSesnEBPyozX-osB3WY,1478
594
+ sglang/srt/layers/quantization/quark/schemes/quark_w4a4_mxfp4.py,sha256=gMKmrpimBzzzQ1_XlqNKHPhJffqoEdifi8EJ6EdCRhE,3758
579
595
  sglang/srt/lora/layers.py,sha256=FaTYf0w3zbKKKAT1nJdAzGRtxvySB6FpNKXvEi5RHH0,12958
580
596
  sglang/srt/lora/lora.py,sha256=sRoU7BdlTV4cnBv1-u3QjisVR9XQrrqAsmhxdNtd8dU,9433
581
597
  sglang/srt/lora/lora_config.py,sha256=qDgMTx_69jyJUl29O5FxLzYa0BMhqYVXWXfyyVOvGm0,1684
582
- sglang/srt/lora/lora_manager.py,sha256=ABuqg1IhY9LNXgJd_v3Yx_VHS52t4SapHmhLiH1thdc,20147
583
- sglang/srt/lora/lora_registry.py,sha256=iBPZFJq2JTa0p1pA_TJ5alU3V95NmazKai4GiDmVXnA,7865
584
- sglang/srt/lora/mem_pool.py,sha256=UYvTgEPvBoVsaX5rEnipebOAlo75pML6_NO9GfCFzl8,12148
598
+ sglang/srt/lora/lora_manager.py,sha256=pwC79GnsTX34qBHJOe12KpSQQrlCqXBUENfqSBhipDQ,22615
599
+ sglang/srt/lora/lora_registry.py,sha256=VBszMm5Hc3WbISONKUmTCTkW7kd0LgDKN67DLvjFVgU,7912
600
+ sglang/srt/lora/mem_pool.py,sha256=yqCAgD2R7G0YwnRTl8h2wq54IExZq0PFgfzipOdEhhM,13274
585
601
  sglang/srt/lora/utils.py,sha256=5G0KiDEn-Zg4OgIOjWnU2ZdQGNstZswoAog9xU4GVDA,5660
586
602
  sglang/srt/lora/backend/base_backend.py,sha256=EIz8I-GIrdmK4fISw3ENhbJVVITaxKfyLxHXGPU4fPs,5044
587
603
  sglang/srt/lora/backend/flashinfer_backend.py,sha256=el6IAB4kTgDTbwCggmqFuukliyoapN5X6FLksG-4wJ8,4151
@@ -591,41 +607,42 @@ sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=QURCYxHNR8Ls4SQtt3dvdgjvdDVh
591
607
  sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=iz5scFNo2YFTeiM8beWg9Z1oZI-6AM_T1wBMCQ6qp2Q,6485
592
608
  sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=Ksova04wgeGsFqGOXWqJtMYaHgyUYcx8VU42BZQOkVA,5129
593
609
  sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=VqCAFvUtq_l-0RGIkx3W_fzD55QcW2FAcqpvSHOBFp4,5115
594
- sglang/srt/managers/cache_controller.py,sha256=sqKPWX0X-q6KjkfvT3UXANGxpG8JvDfm4kLPTC5YvMo,29328
610
+ sglang/srt/managers/cache_controller.py,sha256=umU5ekzYEwt3NL668aGPmtzt1DkttKBjZzAJzl8ipxk,30569
595
611
  sglang/srt/managers/configure_logging.py,sha256=8sNXZ2z9pBWOwn-X3wyz013Ob8Nbm1zDxRkxoZjH-l4,1633
596
612
  sglang/srt/managers/data_parallel_controller.py,sha256=VvEkidmYCGAFTv6upEHf_tyTvkM8xLhCyOnLJGWL6BQ,14592
597
- sglang/srt/managers/detokenizer_manager.py,sha256=SpLxTsSPKBZfD-ZMhJ5zpPPGuUb8PmcYgFSL9CsurU4,10696
598
- sglang/srt/managers/io_struct.py,sha256=hNYPIMEPtDRSHIkxpwD00capl4HAM-64kyIXOUgfk8k,38103
599
- sglang/srt/managers/mm_utils.py,sha256=VrJiv0k_zDa83BsfVv8rmckgXKsvF0Wz88bCAb6R1as,29601
613
+ sglang/srt/managers/detokenizer_manager.py,sha256=uyt4px35nMvPN1d3TXzrMPzhoZExFQqXjpgUUbg5_2Y,10711
614
+ sglang/srt/managers/io_struct.py,sha256=pSetzuUXLC7g0NEjqJcBT4UVcotP4AZ1DlFrh0Us76I,38581
615
+ sglang/srt/managers/mm_utils.py,sha256=kbpD-8t-K4kMdg7FSP6DGK9X4PRuoksTQJkEsdPNdZ8,29378
600
616
  sglang/srt/managers/multimodal_processor.py,sha256=mLnGcUHIvjruwoVhqiz4j7RMYMsZnvQMH8JnKCITDlY,1805
601
- sglang/srt/managers/schedule_batch.py,sha256=pHjIJAdfxNxfMB3Be4AV2yq-MaSN-CkA2yhREeHaxH4,76355
617
+ sglang/srt/managers/schedule_batch.py,sha256=NHWtATO2hYGh86Qi3GlCLMjI_FNV51Ayh-CKWzy7J5A,77343
602
618
  sglang/srt/managers/schedule_policy.py,sha256=l_2C8z52pSi-FOQpF_aR4joj9UL5wP9A0oJlIzSq6bM,22449
603
- sglang/srt/managers/scheduler.py,sha256=oin_9RdC7jiJ-s8XkvuH5HZRSNQojjIsK70t4TyIaFc,106152
619
+ sglang/srt/managers/scheduler.py,sha256=PMwhKXnzRUPvYjjJuv1ME0x26sxTm43YQhT449d9De8,106835
604
620
  sglang/srt/managers/scheduler_input_blocker.py,sha256=zP8xU_UmU2H0AB6sEqvivDcDB1QDgTauNDYDIJ7Nez4,3683
605
621
  sglang/srt/managers/scheduler_metrics_mixin.py,sha256=-1IOE30eOoTvvpz-Xg9J4UcamwAJle2-odFJ9vrinLE,9341
606
- sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=Ei-9lyfrSQAlL9_ULuJLAhA7qpvUzjsBDzBBTDdnt4Q,31394
622
+ sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=e_l5TRm05liINSAeZU-YPRpJXcutmEWYV30F1vPaxbs,31345
607
623
  sglang/srt/managers/scheduler_profiler_mixin.py,sha256=GnR1SALuj_JbsENfmpiVgVcJWtGQn8hJ8TERkLB7h1c,10701
608
- sglang/srt/managers/scheduler_update_weights_mixin.py,sha256=cQYdfHSgqLR7FeWMZl1V_lXMSKw99WGUa9gdiYUMXLU,5334
624
+ sglang/srt/managers/scheduler_recv_skipper.py,sha256=8rVQCBjn9f26Th4XbeUS3fEsvo1GI029Me28wMjcKRY,1103
625
+ sglang/srt/managers/scheduler_update_weights_mixin.py,sha256=R81fU-0aIJoi5URm9IA-n2sEjvn3phGXOhzJALZjiOM,5467
609
626
  sglang/srt/managers/session_controller.py,sha256=dzlMNZlo20FTSl64QqK7y7pElsdCy8ICOWWBPTBVwgs,6040
610
- sglang/srt/managers/template_manager.py,sha256=U79zdED8XSO5Ee9xF13QuPrjI3dVhzAuekRK94DDy_s,10682
611
- sglang/srt/managers/tokenizer_manager.py,sha256=WSMH39L3OxKxbV_WZnBE2e5uUektc21BOzPCX5bBoz8,81929
612
- sglang/srt/managers/tp_worker.py,sha256=Y8GEgEI9Y4ayw4K71yRHDzeVX43JjTBkf3vFf8RfS9w,11764
613
- sglang/srt/managers/tp_worker_overlap_thread.py,sha256=3KQQ2YkyswgtouByhfPPq5mCISsMstn9CEa1i4qv-Xg,10998
627
+ sglang/srt/managers/template_manager.py,sha256=L3eWAhrQoaFYaEKoj9AA7r212bXcj8RJxb9UbhJf9Jg,11823
628
+ sglang/srt/managers/tokenizer_manager.py,sha256=mCisuEEYP-oCY1lfaEl_0xmaYKaZGC09DPSD4cQVSB0,82631
629
+ sglang/srt/managers/tp_worker.py,sha256=923H0m12cVt_b5_6uaBhyceGvbGP56mtOX3l5tGnsnM,11904
630
+ sglang/srt/managers/tp_worker_overlap_thread.py,sha256=NXjrBT7lvdUlDDNFzEQyFYCerpqj92sNl2YynGhQj-Y,11118
614
631
  sglang/srt/managers/utils.py,sha256=RogDdqb2bcxpNk5TbxKyaXFpgFz8t30IOnDp4s-IilM,2776
615
632
  sglang/srt/mem_cache/allocator.py,sha256=id4riucsG9d-P1ikj3-DLJbzhM8-MuJfvDmB8H4g9-o,23364
616
633
  sglang/srt/mem_cache/base_prefix_cache.py,sha256=hLS2ncTMAz7Kpdk5pNwn5c6g8b61_K9OCBL4Vj1xsYc,2801
617
634
  sglang/srt/mem_cache/chunk_cache.py,sha256=FU_fq8e_dodgU8EkRqE4Jr0sKwCXQnKmapO88k6_JfU,3207
618
635
  sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
619
636
  sglang/srt/mem_cache/hicache_storage.py,sha256=USRuV4bvePMxkKr1ypiDnZZSUBbE8Arm5PnZklsbrkc,6385
620
- sglang/srt/mem_cache/hiradix_cache.py,sha256=QbKT2NKLuVe-1AqEc_njtZHXrq8d6WiNNbLa6Qio9qI,25818
637
+ sglang/srt/mem_cache/hiradix_cache.py,sha256=uZvT2elMJ6DFwMMoM_QoKK1UZmmCEP4DcmrBmeRVEeI,27679
621
638
  sglang/srt/mem_cache/memory_pool.py,sha256=i-QhFrYuaAFcMOG7QH6SJyJ8hFzVRS5fjxwKmae8f94,37064
622
- sglang/srt/mem_cache/memory_pool_host.py,sha256=M5dTu4BiEEiTlD7vtctd_LQrCODS1iradGaFtHaRkXM,25385
623
- sglang/srt/mem_cache/multimodal_cache.py,sha256=wZl2KeEl3xeoEsYdH33UoM-FO8kqfLo_XUgereJVvoM,1348
639
+ sglang/srt/mem_cache/memory_pool_host.py,sha256=jbAl2wiYBLkLDam6g-sRlRu1ORmpp5YoEkZ_niMc-mo,25383
640
+ sglang/srt/mem_cache/multimodal_cache.py,sha256=zPnQLQhBZ6zsUpCQPSoNkrB9EEvpoDQS4mU7c3sRWjE,2171
624
641
  sglang/srt/mem_cache/radix_cache.py,sha256=kp4JiVxh3eS4MsFxGKIYjYAb14B777qc1bHrABPqBMc,19101
625
642
  sglang/srt/mem_cache/radix_cache_cpp.py,sha256=1V-XxYC2jxQYUMJBLrlcSMcqhCBo8TC7wRTTVbOaMkY,9269
626
643
  sglang/srt/mem_cache/swa_radix_cache.py,sha256=4qt_gS7GuHiekd6VvKkx54dB2GHJzNA_kj71Np6jTDA,40300
627
644
  sglang/srt/mem_cache/cpp_radix_tree/radix_tree.py,sha256=tQZpz-H6HxNAQe9mYqt6aX8mfDZyu_fbJTcCNK-Ns1M,7179
628
- sglang/srt/mem_cache/storage/hf3fs/client_hf3fs.py,sha256=k9ZtqKfWdJh19jLyRu0iXJEciXXuqFcEuMXej3hJhmg,5178
645
+ sglang/srt/mem_cache/storage/hf3fs/client_hf3fs.py,sha256=WV5M9UPPLUMMjU3Mpp6HBqyCQpGBuMyj-N5PwQcNo5k,5178
629
646
  sglang/srt/mem_cache/storage/hf3fs/hf3fs_utils.cpp,sha256=dDYPtxmL6OFhEyv_GuTgzkTDpBjcD0EK8kphnn7eaGc,1133
630
647
  sglang/srt/mem_cache/storage/hf3fs/storage_hf3fs.py,sha256=H0X4v5AWs4uTHoBGgb9BgfKf3WZ3reZKpLNxmL1dTbI,9230
631
648
  sglang/srt/mem_cache/storage/hf3fs/test_hf3fs_utils.py,sha256=g2h0Woa2CK39pRTigHCVR_hq095dTaC70lCIAIs9THM,1089
@@ -636,15 +653,16 @@ sglang/srt/mem_cache/storage/nixl/nixl_utils.py,sha256=i2C70Ct2xdZtR_6OSL7uQk7pe
636
653
  sglang/srt/mem_cache/storage/nixl/test_hicache_nixl_storage.py,sha256=TBtN12muLFXZ296mK2d3TSTyj8qhMo5kAYWLpSjLBhY,8017
637
654
  sglang/srt/metrics/collector.py,sha256=FH6Wil2lMK7d7q74CfxfdULCaHKDXeAoo8UBZqqdepM,20289
638
655
  sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
639
- sglang/srt/model_executor/cuda_graph_runner.py,sha256=DAJwRTRqxm-D2Mp-ZhshL_WjMQTg0IKI5f1Jht4FO-o,33281
640
- sglang/srt/model_executor/forward_batch_info.py,sha256=Drc3TsN5XJNkaRiJXHmCTwIazXsIEfmNi6TdQfBpJG4,38831
641
- sglang/srt/model_executor/model_runner.py,sha256=4tWBilTr2ONfwsLuQZEkXK9qtxHyUcVxaJOwxmM6IYw,73387
656
+ sglang/srt/model_executor/cuda_graph_runner.py,sha256=47ppwENzKmm8NuTlNPY9H3oWhbBHytVc4txQpdowUZA,33303
657
+ sglang/srt/model_executor/forward_batch_info.py,sha256=CFMEfynC-lpIrFfllHgINqsZwr7qqyBIgbPc8Ypcoyw,39400
658
+ sglang/srt/model_executor/model_runner.py,sha256=Z6WXH_PaU2AZZcEoXDQYNgrSOLu9s5kFxURkcedZcGs,74005
642
659
  sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
643
660
  sglang/srt/model_loader/loader.py,sha256=wxSUYrqv23w9yHYy9Lm0ozWggs9pf6Co4yqe1UO8PZk,59799
644
661
  sglang/srt/model_loader/utils.py,sha256=svSrPQxViBz_-0_sByu4coCYrzJ51-4FCi_bFqa6rDo,4474
645
- sglang/srt/model_loader/weight_utils.py,sha256=2UxbNYzoZxIIrDhsEOk6M-ofFg5s1UMtV1j23XRbnq4,38129
662
+ sglang/srt/model_loader/weight_utils.py,sha256=dHEkm0c4w-lyhJBzktvmop-Psmkpb2KoTu7Azgw7VgU,38555
646
663
  sglang/srt/models/arcee.py,sha256=Q6fujRESEnljzs8ls3XGsPtWxzfKfUa0QAvAekYzeBY,18981
647
664
  sglang/srt/models/baichuan.py,sha256=HbvlErnkCSK4pRQYCSDxMcrn-1DQyfiNoeDcnRrJas8,15807
665
+ sglang/srt/models/bailing_moe.py,sha256=1I5xcwtS-y0EOKje7q3j6ORIj8gBuasHS0WmmOCRHIc,14395
648
666
  sglang/srt/models/bert.py,sha256=ODJe8YfNRP-hHsomFWk4_QpcuiSsNfjzGf256EDS0Pc,15802
649
667
  sglang/srt/models/chatglm.py,sha256=cajLN9caBl09e0TwOFkiTTKDqwlbmHo_yS-NCjdeQW8,13957
650
668
  sglang/srt/models/clip.py,sha256=kOPSXnmgH_QKu2-ENn49HLDS9yf7Gm1HCOE_YvoNhlA,19858
@@ -653,8 +671,10 @@ sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,155
653
671
  sglang/srt/models/deepseek.py,sha256=sqobTr9a9c5pNLNJZE-WrJIE_qBq0lC5gCGv9NpGU_Q,17364
654
672
  sglang/srt/models/deepseek_janus_pro.py,sha256=td8xGs6ARfJ8AQCYwUhMOZoWigrAs7m3trF5-kXCqik,70418
655
673
  sglang/srt/models/deepseek_nextn.py,sha256=47fehxRdiOizr0rdLg5f1fzQEx6gGAOcDcWKtblloyk,5928
656
- sglang/srt/models/deepseek_v2.py,sha256=e28nSn6NJq9YlsbJWPqotIKBJr66Z9XyBPBk44K0azw,104084
674
+ sglang/srt/models/deepseek_v2.py,sha256=PWE4zFCHr3SThN3mh2ktiKZJHRSWHgSSyHjCQfOgmHI,105516
657
675
  sglang/srt/models/deepseek_vl2.py,sha256=7X3gI4DE7guUtNJvMLf06jO8MUHKa2Aan8evZg5hsXc,13061
676
+ sglang/srt/models/ernie4.py,sha256=_s_Cvw-lwoeSAvGBHSfG1X8_pmPJoJv6e6TvAJGYfoA,16084
677
+ sglang/srt/models/ernie4_eagle.py,sha256=E7f-ygCcY8pIoyTMKFyVC2uRcG59-_RaQTO4gd5_uWI,7223
658
678
  sglang/srt/models/exaone.py,sha256=TpO-rtCpEZ8Ua7hGFnS8l2oAYhY0Pij50grc9WQ2mvc,13576
659
679
  sglang/srt/models/gemma.py,sha256=2atKwFN5wCsJCmBRmURX9vaCcFjxlFirL7xsnG5KPlI,14180
660
680
  sglang/srt/models/gemma2.py,sha256=lpHKN-ovN9IWz3iA9FsR5H86lGatectnM8sjx0GoinM,18033
@@ -663,12 +683,13 @@ sglang/srt/models/gemma3_causal.py,sha256=0FdI0wVoJ4ZakBl8ShO29D1uEMO0T8CRjdsfiH
663
683
  sglang/srt/models/gemma3_mm.py,sha256=sRvdAOiStlDhC_-sSH226d0awx4ZQJG7Wtdf7PE6ygw,17189
664
684
  sglang/srt/models/gemma3n_audio.py,sha256=isgKfjA5UieYawxU6medL2ssXlzYPqAbagDBnLcemC0,36405
665
685
  sglang/srt/models/gemma3n_causal.py,sha256=kBDlpSWAg6ZDmUPe08Q7sj5CbdVD_0h07zwJw8sKu7I,36335
666
- sglang/srt/models/gemma3n_mm.py,sha256=pL_fvnkz9fKaC2Afms5KG3Me3u5sp1C2-mAUhqPGrQ4,18701
686
+ sglang/srt/models/gemma3n_mm.py,sha256=TpyhU0Q_T-pZHc5_oo_o7WDVv5kW6cvLlOqRF8AoUng,20439
667
687
  sglang/srt/models/glm4.py,sha256=2VQzUqFkQTy_2nfkxP9SF6_9kKLTZUExGRjge7r99Es,11265
668
- sglang/srt/models/glm4_moe.py,sha256=CXXPvglPPWWSmeEkf2x5iezp_-tBWY_dcDKqG6v6zzg,42890
688
+ sglang/srt/models/glm4_moe.py,sha256=6y3VVwL313JpA7CED65-xkuMhqJPrgL_uzrb8AwthwQ,43215
669
689
  sglang/srt/models/glm4_moe_nextn.py,sha256=cWB6o3w7xBvgi2ouGKsqZ3iovKH65QiRQul42wcgzMk,5897
670
690
  sglang/srt/models/gpt2.py,sha256=kclhxEs8oJk1KCyhmAqo7rZqecVGGHYkc-a1WZi3aIk,9841
671
691
  sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtYhl4,10307
692
+ sglang/srt/models/gpt_oss.py,sha256=fuKmLBTkesmAOnx8IWJI24mvSn7Hhfs-GtGu73F37z0,42929
672
693
  sglang/srt/models/granite.py,sha256=5WOJyNYAlt5RNHSexNfPNihhSxIMd7wPzju1cTixKig,20852
673
694
  sglang/srt/models/granitemoe.py,sha256=Yr1pEOS_8QopQKBIiDpFpMXZ0Wkc1REyPorM3eDTjNE,13808
674
695
  sglang/srt/models/grok.py,sha256=M23KdbuB3rrJoOG2LkS6EvVts1p0fMfmYAzcbjXQy-Q,28243
@@ -712,23 +733,23 @@ sglang/srt/models/phi4mm_utils.py,sha256=bW3un4x3SHDvSkAOHQEFO286FGdOPeKhBpJCOmN
712
733
  sglang/srt/models/phimoe.py,sha256=EGaKFi02MZP20qFrILA4o2SJ0wb1HoJagFTSqVUbJcQ,19223
713
734
  sglang/srt/models/pixtral.py,sha256=ZkZrV0mEfmbBdgRko2AvsMFZbAVskgp_5a5p1xCSRN0,16835
714
735
  sglang/srt/models/qwen.py,sha256=j0OU3JjznWi_TbayI09yYGM-Obov6mQzJO82Hs0qBGw,11868
715
- sglang/srt/models/qwen2.py,sha256=0gnggs2q-_FtD75J-UM4KMNBXSapVQGz73B-C2yZHDg,22437
736
+ sglang/srt/models/qwen2.py,sha256=hUM8r5SJLOWRfTq9LuFMuoAkP81B0qFi8r23tvrnG2I,22758
716
737
  sglang/srt/models/qwen2_5_vl.py,sha256=vOuIyK_NLuGgz9LlAdyM7cSQUU93VnBCYXUr_n4iOLs,22708
717
738
  sglang/srt/models/qwen2_audio.py,sha256=lVeYmC29eVZIR9I6ZjFe4xvC0wSXR0NtGCli6fSXtms,7870
718
739
  sglang/srt/models/qwen2_classification.py,sha256=dGrMm4ebd30_lBhHOhaV57ig2iOTx3nqB4GEzsrRIM8,2747
719
740
  sglang/srt/models/qwen2_eagle.py,sha256=6spFg6JYQGeUXLv2Q5eMkVzZvnqGyvW6SdVBtFMxlTM,4806
720
- sglang/srt/models/qwen2_moe.py,sha256=Mgwe-yg22rkPOOic_ewkfb7C9DEQsx4irtUO9GXG4Vo,25664
741
+ sglang/srt/models/qwen2_moe.py,sha256=LPiuiQbAcoZwM17OR0_Jf5ddPRc8PORzVjhknOZQ6f8,25985
721
742
  sglang/srt/models/qwen2_rm.py,sha256=-mQXDEv11p-I1HXgYLTtY6ROem6UYorO958WsDrzsgs,2837
722
743
  sglang/srt/models/qwen2_vl.py,sha256=ZRAL32Ymt4ZCBqB0JorXAm4ZBrECZ9EbaXKaTSikkQM,21366
723
744
  sglang/srt/models/qwen3.py,sha256=COSC1TsCQNTq3E1QoVJYj1yTn-CFEVEX7keRwrrBXLs,20121
724
- sglang/srt/models/qwen3_moe.py,sha256=C3zzu50iHEbTb5KUZp9I0672m6bruR0lShz4I5m19Lo,31434
745
+ sglang/srt/models/qwen3_moe.py,sha256=MUHbQ52QoQgslfak2gedOnhdZFcYo2f2LHvxihsZ1k4,32728
725
746
  sglang/srt/models/registry.py,sha256=bH9H0OiNV0Cn7FRBM30Oh6dxICTs9mJscdLtHYPJvxc,3783
726
747
  sglang/srt/models/roberta.py,sha256=3k53V2Gbezk3jU-D03thXx8csGn5DxFK01ZQ9WFiWPI,9828
727
748
  sglang/srt/models/siglip.py,sha256=DIVJRwdtpLD2QT1kVPIHw7Bn9BE40xJbkHujDNzkjgA,9892
728
749
  sglang/srt/models/stablelm.py,sha256=0x_31uIr3WcWwecdPAI3ek9KkyKBJS7VwknTk2y0gjY,12281
729
- sglang/srt/models/step3_vl.py,sha256=j7EdJN6zSsFkDAz9tRGTpKFMw6XJueJEIdFrTVZoALk,35138
750
+ sglang/srt/models/step3_vl.py,sha256=MrYg9L3BbSj8fwRzipeSUOCM0tmqibip6qmEqjTYJRg,35600
730
751
  sglang/srt/models/torch_native_llama.py,sha256=oTxxof5Wx2b2egIwWx56LeHjtPGQleQ2zM5jXHMiuOU,19293
731
- sglang/srt/models/transformers.py,sha256=_CL0npi-_HvqzxkLgFhv2EE3bfGvCLXPYAtPFywZOr0,10260
752
+ sglang/srt/models/transformers.py,sha256=ChqRbUq5YD9oXnXwVP81XRoZmsKc44ouYDNgRsH6zmk,10249
732
753
  sglang/srt/models/vila.py,sha256=3NiRHWOqXnCRjXmswAldsNtiE5zRILbScEsbhUzLZF4,9846
733
754
  sglang/srt/models/xverse.py,sha256=DsNVI9JpzN4jj0Ry6aTrj7r-xq5YLOoDX2kH4YLJA-I,14035
734
755
  sglang/srt/models/xverse_moe.py,sha256=7KCM2-j12towDMNvXkuuYiBOmNauH6NG4Ip40x0khqA,16782
@@ -750,7 +771,7 @@ sglang/srt/multimodal/processors/phi4mm.py,sha256=8GaUVpGucV4UKoa0XkxN0gDpMZax6U
750
771
  sglang/srt/multimodal/processors/pixtral.py,sha256=Lx5Z-9NkovLDTHlRBznHxGn4r2XhDz8am2XpOskOVQY,3203
751
772
  sglang/srt/multimodal/processors/qwen_audio.py,sha256=jxtArfNuBH4q9LQVEydMDsj85avqBGNl-v1ZZTjC9E0,2367
752
773
  sglang/srt/multimodal/processors/qwen_vl.py,sha256=tZOpz8dabyNvjS4hn0UxLf-EaEZH8kmHjyZaYRm9EZ8,10498
753
- sglang/srt/multimodal/processors/step3_vl.py,sha256=JEHsj9SQw5udc3s-o1rwBF-WsrMTCxUcbgM4qFXKTT0,18240
774
+ sglang/srt/multimodal/processors/step3_vl.py,sha256=9OI6UW9ybGbAnYN_S-5jh7TN8BGM_bQkvh1AHBJKzAQ,18350
754
775
  sglang/srt/multimodal/processors/vila.py,sha256=vWgV1lTdjiFiTR8w-7fC66-L_tpOOD59c-c9NljGelE,2192
755
776
  sglang/srt/sampling/custom_logit_processor.py,sha256=AwODYVJdRkcQ8PGtJrhzKsqAgn8XZLQbAmR9fGiQzmc,1608
756
777
  sglang/srt/sampling/sampling_batch_info.py,sha256=Duk4Fp99QZyOFld3i4AGwkW9FlULY86CNKv6v0_kA4w,14683
@@ -771,7 +792,7 @@ sglang/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
771
792
  sglang/test/few_shot_gsm8k.py,sha256=qnEejCyPHGEMBEiNc7T4-t9lvoo2DNN72RXw-ei_TnY,4300
772
793
  sglang/test/few_shot_gsm8k_engine.py,sha256=Yf1Dg0Tk-tY4v1oj_UTxJ7950CWn6kRMbQKDjT2Orgo,3902
773
794
  sglang/test/run_eval.py,sha256=K4Z3OdMPvLeuGJhWWTBD0F-MOUFeBYLvsPnDQ44bd18,4036
774
- sglang/test/runners.py,sha256=5kCwKrww_TQQSoSV3AMnWPl_qG_l9tH-OKQEgxwAxJQ,33731
795
+ sglang/test/runners.py,sha256=a8Edv5sSo8EeEEiOrKbPrCS8Z0ADaRzfSpdtzG6b_qM,33761
775
796
  sglang/test/send_one.py,sha256=XNZdfZs21tXDBwrDj2pq46WQd5smiKoYETY8U09c1Z8,5079
776
797
  sglang/test/simple_eval_common.py,sha256=OW4Z7O33QqDySRFuHIC4Ouidfup4eaOEuR9i7xJzkDY,12379
777
798
  sglang/test/simple_eval_gpqa.py,sha256=8Xt9Bw05c7SZTYrCZgB68OZUqUbLo69ywiyx0bTvSUk,3220
@@ -793,14 +814,14 @@ sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg
793
814
  sglang/test/test_marlin_moe.py,sha256=PITMGcGF4ci2csXJejdLSNuxpURnFmcflsipAZsGakU,7991
794
815
  sglang/test/test_marlin_utils.py,sha256=gFMJwxbfEEJDeHI-HU5FDF9FFqqcM-yahK0TWyVIVgQ,5171
795
816
  sglang/test/test_programs.py,sha256=DJIbTChloPwaDiEWHQoSTuKv3BGx0-K8Y5e1mDDTACM,18910
796
- sglang/test/test_utils.py,sha256=cCvqi0eLPaOK41a351iji4g1cVmm6VRpkpuGSdrZF28,41131
817
+ sglang/test/test_utils.py,sha256=2vUWNlGaInhG5iLNIUOmULlO65CLzojRpbX2yZamj9w,41155
797
818
  sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
798
819
  sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
799
820
  sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
800
821
  sglang/test/attention/test_prefix_chunk_info.py,sha256=hpoDe2wfSa6RlUbfyri_c0iyBTb35UXGL9I2Xh6jamM,7772
801
822
  sglang/test/attention/test_trtllm_mla_backend.py,sha256=D7KOJbsfGRSdO5dMlGokBpiC6Sh42g51HlVkgxSK-yI,36251
802
- sglang-0.4.10.post2.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
803
- sglang-0.4.10.post2.dist-info/METADATA,sha256=j--rk52lslAMmr3aXQyGrZ9W8nJa9RurXTh8ENc_twQ,27433
804
- sglang-0.4.10.post2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
805
- sglang-0.4.10.post2.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
806
- sglang-0.4.10.post2.dist-info/RECORD,,
823
+ sglang-0.5.0rc0.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
824
+ sglang-0.5.0rc0.dist-info/METADATA,sha256=rGKDl9lYZ-foVU7hoyNLPRsDGmbsI_lB_NWUtjJKJd8,27541
825
+ sglang-0.5.0rc0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
826
+ sglang-0.5.0rc0.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
827
+ sglang-0.5.0rc0.dist-info/RECORD,,