sglang 0.4.7.post1__py3-none-any.whl → 0.4.8.post1__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 (123) hide show
  1. sglang/bench_one_batch.py +8 -6
  2. sglang/srt/_custom_ops.py +2 -2
  3. sglang/srt/code_completion_parser.py +2 -44
  4. sglang/srt/configs/model_config.py +1 -0
  5. sglang/srt/constants.py +3 -0
  6. sglang/srt/conversation.py +14 -3
  7. sglang/srt/custom_op.py +11 -1
  8. sglang/srt/disaggregation/base/conn.py +2 -0
  9. sglang/srt/disaggregation/decode.py +22 -28
  10. sglang/srt/disaggregation/decode_schedule_batch_mixin.py +4 -3
  11. sglang/srt/disaggregation/mini_lb.py +34 -4
  12. sglang/srt/disaggregation/mooncake/conn.py +301 -64
  13. sglang/srt/disaggregation/mooncake/transfer_engine.py +31 -1
  14. sglang/srt/disaggregation/nixl/conn.py +94 -46
  15. sglang/srt/disaggregation/prefill.py +20 -15
  16. sglang/srt/disaggregation/utils.py +47 -18
  17. sglang/srt/distributed/parallel_state.py +12 -4
  18. sglang/srt/entrypoints/engine.py +27 -31
  19. sglang/srt/entrypoints/http_server.py +149 -79
  20. sglang/srt/entrypoints/http_server_engine.py +0 -3
  21. sglang/srt/entrypoints/openai/__init__.py +0 -0
  22. sglang/srt/{openai_api → entrypoints/openai}/protocol.py +115 -34
  23. sglang/srt/entrypoints/openai/serving_base.py +149 -0
  24. sglang/srt/entrypoints/openai/serving_chat.py +897 -0
  25. sglang/srt/entrypoints/openai/serving_completions.py +425 -0
  26. sglang/srt/entrypoints/openai/serving_embedding.py +170 -0
  27. sglang/srt/entrypoints/openai/serving_rerank.py +102 -0
  28. sglang/srt/entrypoints/openai/serving_score.py +61 -0
  29. sglang/srt/entrypoints/openai/usage_processor.py +81 -0
  30. sglang/srt/entrypoints/openai/utils.py +72 -0
  31. sglang/srt/function_call/base_format_detector.py +7 -4
  32. sglang/srt/function_call/deepseekv3_detector.py +1 -1
  33. sglang/srt/function_call/ebnf_composer.py +64 -10
  34. sglang/srt/function_call/function_call_parser.py +6 -6
  35. sglang/srt/function_call/llama32_detector.py +1 -1
  36. sglang/srt/function_call/mistral_detector.py +1 -1
  37. sglang/srt/function_call/pythonic_detector.py +1 -1
  38. sglang/srt/function_call/qwen25_detector.py +1 -1
  39. sglang/srt/{openai_api/utils.py → jinja_template_utils.py} +6 -5
  40. sglang/srt/layers/activation.py +28 -3
  41. sglang/srt/layers/attention/aiter_backend.py +5 -2
  42. sglang/srt/layers/attention/base_attn_backend.py +1 -1
  43. sglang/srt/layers/attention/cutlass_mla_backend.py +1 -0
  44. sglang/srt/layers/attention/flashattention_backend.py +43 -23
  45. sglang/srt/layers/attention/flashinfer_backend.py +9 -6
  46. sglang/srt/layers/attention/flashinfer_mla_backend.py +7 -4
  47. sglang/srt/layers/attention/flashmla_backend.py +5 -2
  48. sglang/srt/layers/attention/tbo_backend.py +3 -3
  49. sglang/srt/layers/attention/triton_backend.py +19 -11
  50. sglang/srt/layers/communicator.py +5 -5
  51. sglang/srt/layers/dp_attention.py +11 -2
  52. sglang/srt/layers/layernorm.py +44 -2
  53. sglang/srt/layers/linear.py +18 -1
  54. sglang/srt/layers/logits_processor.py +14 -5
  55. sglang/srt/layers/moe/ep_moe/kernels.py +159 -2
  56. sglang/srt/layers/moe/ep_moe/layer.py +286 -13
  57. sglang/srt/layers/moe/ep_moe/token_dispatcher.py +19 -2
  58. sglang/srt/layers/moe/fused_moe_native.py +7 -0
  59. sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=384,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  60. sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +13 -2
  61. sglang/srt/layers/moe/fused_moe_triton/layer.py +148 -26
  62. sglang/srt/layers/moe/topk.py +117 -4
  63. sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py +6 -2
  64. sglang/srt/layers/quantization/fp8.py +25 -17
  65. sglang/srt/layers/quantization/fp8_utils.py +5 -4
  66. sglang/srt/layers/quantization/modelopt_quant.py +62 -8
  67. sglang/srt/layers/quantization/utils.py +5 -2
  68. sglang/srt/layers/rotary_embedding.py +144 -12
  69. sglang/srt/layers/sampler.py +1 -1
  70. sglang/srt/layers/vocab_parallel_embedding.py +14 -1
  71. sglang/srt/lora/lora_manager.py +173 -74
  72. sglang/srt/lora/mem_pool.py +49 -45
  73. sglang/srt/lora/utils.py +1 -1
  74. sglang/srt/managers/cache_controller.py +33 -15
  75. sglang/srt/managers/expert_distribution.py +21 -0
  76. sglang/srt/managers/io_struct.py +19 -14
  77. sglang/srt/managers/multimodal_processors/base_processor.py +44 -9
  78. sglang/srt/managers/multimodal_processors/gemma3n.py +97 -0
  79. sglang/srt/managers/schedule_batch.py +49 -32
  80. sglang/srt/managers/schedule_policy.py +70 -56
  81. sglang/srt/managers/scheduler.py +189 -68
  82. sglang/srt/managers/template_manager.py +226 -0
  83. sglang/srt/managers/tokenizer_manager.py +11 -8
  84. sglang/srt/managers/tp_worker.py +12 -2
  85. sglang/srt/managers/tp_worker_overlap_thread.py +11 -0
  86. sglang/srt/mem_cache/{paged_allocator.py → allocator.py} +125 -34
  87. sglang/srt/mem_cache/base_prefix_cache.py +52 -8
  88. sglang/srt/mem_cache/chunk_cache.py +11 -16
  89. sglang/srt/mem_cache/hiradix_cache.py +34 -23
  90. sglang/srt/mem_cache/memory_pool.py +118 -114
  91. sglang/srt/mem_cache/radix_cache.py +20 -16
  92. sglang/srt/model_executor/cuda_graph_runner.py +77 -46
  93. sglang/srt/model_executor/forward_batch_info.py +18 -5
  94. sglang/srt/model_executor/model_runner.py +27 -8
  95. sglang/srt/model_loader/loader.py +50 -8
  96. sglang/srt/model_loader/weight_utils.py +100 -2
  97. sglang/srt/models/deepseek_nextn.py +35 -30
  98. sglang/srt/models/deepseek_v2.py +255 -30
  99. sglang/srt/models/gemma3n_audio.py +949 -0
  100. sglang/srt/models/gemma3n_causal.py +1009 -0
  101. sglang/srt/models/gemma3n_mm.py +511 -0
  102. sglang/srt/models/glm4.py +312 -0
  103. sglang/srt/models/hunyuan.py +771 -0
  104. sglang/srt/models/mimo_mtp.py +2 -18
  105. sglang/srt/reasoning_parser.py +21 -11
  106. sglang/srt/server_args.py +51 -9
  107. sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +131 -10
  108. sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py +125 -12
  109. sglang/srt/speculative/eagle_utils.py +80 -8
  110. sglang/srt/speculative/eagle_worker.py +124 -41
  111. sglang/srt/torch_memory_saver_adapter.py +19 -15
  112. sglang/srt/two_batch_overlap.py +4 -1
  113. sglang/srt/utils.py +248 -11
  114. sglang/test/test_block_fp8_ep.py +1 -0
  115. sglang/test/test_utils.py +1 -0
  116. sglang/version.py +1 -1
  117. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.post1.dist-info}/METADATA +4 -10
  118. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.post1.dist-info}/RECORD +121 -105
  119. sglang/srt/entrypoints/verl_engine.py +0 -179
  120. sglang/srt/openai_api/adapter.py +0 -2148
  121. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.post1.dist-info}/WHEEL +0 -0
  122. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.post1.dist-info}/licenses/LICENSE +0 -0
  123. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.post1.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=TwgXZYmwPaHVsdPtNU9LO0p1tr5OOKLy9wYgrfGAlFU,14056
4
- sglang/bench_one_batch.py,sha256=fCxRRk14uQqTxpriBkrUkvrOUjNT7TZgDeGGWJQzhgw,19328
4
+ sglang/bench_one_batch.py,sha256=RJsXQ7pvq8y4yzW7svURtWZF5RiawxEMrsjV0OcKsrk,19388
5
5
  sglang/bench_one_batch_server.py,sha256=4Jbs8UI-Sbt3TL3SJBC3fySZPe4apz2glKWOvvaYsQQ,13486
6
6
  sglang/bench_serving.py,sha256=EzZn91DuScAEhUMQv-LPWm3UoGjzj8OEzHMzVwoEA34,65672
7
7
  sglang/check_env.py,sha256=qDMIG2rCNBH1yKnxQmF-Bp10oiFMUKMgfZLHZYOmdSY,8412
@@ -11,7 +11,7 @@ sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
11
11
  sglang/math_utils.py,sha256=QYtbaIA76P33ojcOrL32VR6yXWv-Od_3WCZNO4kQ-YQ,177
12
12
  sglang/profiler.py,sha256=tEHzHerXC-ymk4OrkoUcMbgcGHmb8VESthsNSP2Yx9w,4417
13
13
  sglang/utils.py,sha256=VH6zrnkjzcR3DE__WfVph6wswJ4JuzoQD47VmbZ38eI,16435
14
- sglang/version.py,sha256=rkzrlJTNQdK78CIF17Jk3y_9tOj5bACvLDEnSq4qJRw,28
14
+ sglang/version.py,sha256=E6iPG1WE6yyF-fToZYo4ZM-iwegVLaXOKuEXlAJ_kvg,28
15
15
  sglang/eval/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
16
16
  sglang/eval/loogle_eval.py,sha256=pRPVA4fxGmT3_oXvXnlNE-UlPrcQGLBJF-OSE9YWJXM,4336
17
17
  sglang/lang/chat_template.py,sha256=HKlx7snSWFED8GKF5ex79sQrPWFw5TSXQM0_LsiD9Bc,20552
@@ -27,23 +27,25 @@ sglang/lang/backend/litellm.py,sha256=ugmL7sfUxkUHVbHtwNzHgdQAEd4UCjNQboFuE3KThc
27
27
  sglang/lang/backend/openai.py,sha256=YFs7pCDK6_DoYmjTa-vgDVdaeE4PYr7E-sIC1q7vAns,16422
28
28
  sglang/lang/backend/runtime_endpoint.py,sha256=oxLCLXCa26VfKyPV-wbRjLrQvqTG8NzoaVgwczzyIYc,17481
29
29
  sglang/lang/backend/vertexai.py,sha256=gz0uNYyBb88jbPYz6ZIJ774fefrcbuVdoK33bphUZpI,4827
30
- sglang/srt/_custom_ops.py,sha256=Y56OZL-6cUBLVmCpWSSjmpsisRYbSE9H--Oq8NjzbPM,4437
30
+ sglang/srt/_custom_ops.py,sha256=0lJRMTKTjoxJPh1qQnnMY02Z3SyBDi7LJI34IBLQsgQ,4462
31
31
  sglang/srt/aio_rwlock.py,sha256=6LYtOdeTUY3hkfa1dmYkgsaF2ttrwIF3hUWz2AZ2fqw,2970
32
- sglang/srt/code_completion_parser.py,sha256=vafJqdTTreD9kvCsoMuxP_fFmoue7pOA1ibI9KXg4mE,5405
33
- sglang/srt/conversation.py,sha256=Ep72tO6lJ7vpyhCmzp5AGmrjDwc2VkegLs2gIklGm4g,36628
34
- sglang/srt/custom_op.py,sha256=lIwo6blTAaPUqXnxy1xsmUJ0qCdp9poPKbZ4mRJAaN4,2734
32
+ sglang/srt/code_completion_parser.py,sha256=KFa95OU0TeVjJkOqIgS4xV3kaJ5dFWOmAAgISyc1oEc,3803
33
+ sglang/srt/constants.py,sha256=0i-tEwG2BSYNDy96MxnGHV5HnBELkYcnsVGsE-R18o0,93
34
+ sglang/srt/conversation.py,sha256=xkV_OWdotT_Tf1QzpxLL-oZ-THAKKtJj49Q9B_L-WT8,37144
35
+ sglang/srt/custom_op.py,sha256=87r2PIgiGLREsIZQ8qsUD-zgI66_54y9GrE0buXzoCI,3076
35
36
  sglang/srt/debug_utils.py,sha256=slaFOY4BYDBFatkfu8FZlzai-u4LFS-5GUzdr-t50zE,2241
36
37
  sglang/srt/hf_transformers_utils.py,sha256=S1ZF4aFKTocenXvj9ti6M-buiCBus4FrP1AdwFz_Wbw,11943
38
+ sglang/srt/jinja_template_utils.py,sha256=Jc6Vl-lYtUyMBSWWQGIMpJQ1O-ceU9c8DbOi_bN-oVk,6719
37
39
  sglang/srt/mm_utils.py,sha256=7o4w1W0VAWgEquVIz7qw7-IxYbZY2g3MrXw9php90oY,12621
38
40
  sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
39
41
  sglang/srt/operations.py,sha256=ddQ8KO63L73OciaR8MZ9h2h83gKVY4-WuWgeEGowPJA,5346
40
42
  sglang/srt/operations_strategy.py,sha256=6DDLEdmkLrFDTXWZoFO0q9MZjEThvfyvoO-LbQsNpPQ,7023
41
43
  sglang/srt/patch_torch.py,sha256=OUPCGQSQz3MVZB1zZ_Eq8lXiw0uIKJ_HWjqQolI8FsM,3088
42
- sglang/srt/reasoning_parser.py,sha256=lZN5b1TJp3EK2GzxcICQhp0tRQjQw3ROOrjRgQe6Bsc,6341
43
- sglang/srt/server_args.py,sha256=nz73pQa4Z6mI-2elebYMV0ajq3QLH1dQ3OBslp-G7wU,71676
44
- sglang/srt/torch_memory_saver_adapter.py,sha256=HYlgYJ2sgmjs2RSjU2KbCaXijRg3mTDZ0ZcCB5Bt6Ps,2211
45
- sglang/srt/two_batch_overlap.py,sha256=Q6tobqItnlK5ad9mqqMMx0HJ-PAzW4DgBtgSQ7Ax6sM,21848
46
- sglang/srt/utils.py,sha256=Zl-xEOFgPyr91A9-q-0Y1PUobAmKvKURgSDR3EBGk4U,72516
44
+ sglang/srt/reasoning_parser.py,sha256=vf0kWBM4IXwbuzGBIOMdiXdn9gavqkHb1QIaTbkU7vc,6742
45
+ sglang/srt/server_args.py,sha256=MVp3qB7PmJlw82hY8dQPoVsJZ26RpmiBqLDgzq1vjOc,73495
46
+ sglang/srt/torch_memory_saver_adapter.py,sha256=K_eTx0UU84MHSTXI3iqYLdHV4IWtJMJ2FKdGFJR8v1E,2417
47
+ sglang/srt/two_batch_overlap.py,sha256=M5Ca3LV4_j7g3o78WQDjWzgsSCpY4E75SIKnwoa7tQg,21922
48
+ sglang/srt/utils.py,sha256=Xg7K6VaTFDrEdW8b07AVREPHdx9M4vJuOI3aIct82JM,80640
47
49
  sglang/srt/warmup.py,sha256=FmJiYfjRr3X_eAe7ojQaPoN17LvHpjDmRWRnO-k86AQ,1469
48
50
  sglang/srt/configs/__init__.py,sha256=8EcVRP95epZ49DxBa6LgKWt7eO3Qe7Hrr3V1c6HkMnY,553
49
51
  sglang/srt/configs/chatglm.py,sha256=j-b0YkdYUmQm2y1kNmMJtKeACxWKmBbvNNkDWbs6kbI,2907
@@ -56,7 +58,7 @@ sglang/srt/configs/janus_pro.py,sha256=Rrb7kQsNaUP-TiZrjNk8Lr1momFrql8ScEunnrH0_
56
58
  sglang/srt/configs/kimi_vl.py,sha256=4W7VQI3pr888ZsFA2SqCQo4mI0seXTOrGQ-x3oTvWew,1358
57
59
  sglang/srt/configs/kimi_vl_moonvit.py,sha256=hx2Rt4JSFbvy2HUTeLjBpge87m8M6ITAhqsgdNf_Jd4,1163
58
60
  sglang/srt/configs/load_config.py,sha256=qs-AxuplouBx2tsv9KGBOLZPbwzuVA4vbktbGP_cRp8,3309
59
- sglang/srt/configs/model_config.py,sha256=mXERUkexcWnZxrCzpxaMo7FhYm-CNhwttDIVhw-ZysY,25206
61
+ sglang/srt/configs/model_config.py,sha256=ESZEDSvxVfRH4CiZoy8JVpapJU5yPTYCAxZB38AJ7b0,25245
60
62
  sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
61
63
  sglang/srt/connector/__init__.py,sha256=czLX5JOxuMhH-T9eSJzoc1qv1B4z9chyffDRL5I6wo4,1247
62
64
  sglang/srt/connector/base_connector.py,sha256=i6i1TIzsz4NbSEkrdMPq-urb2sN2aLAx8dazga4gB9U,2833
@@ -73,28 +75,28 @@ sglang/srt/constrained/outlines_jump_forward.py,sha256=Gyubp-FVetxd6wP4FA_kD6cCX
73
75
  sglang/srt/constrained/reasoner_grammar_backend.py,sha256=YFxWuOTTo4e6cGhxnaBwuwli6f8FTUJtszib8dBq_8I,3207
74
76
  sglang/srt/constrained/xgrammar_backend.py,sha256=u9Ao-XX22UzALkATEAR_-0EoyBTxMTJV590qd9LAUHM,7906
75
77
  sglang/srt/constrained/triton_ops/bitmask_ops.py,sha256=WjTen9iuuFWLzkE1mAHQZB9_7aIy5QH8Wjf-lB-Fams,4614
76
- sglang/srt/disaggregation/decode.py,sha256=vkxzLt843pcUe9gQI2M--25fyzCKpi0iGrLa7QPKFhA,33128
77
- sglang/srt/disaggregation/decode_schedule_batch_mixin.py,sha256=cUfd3-5lRHqWpwkrj4xdGMV2Hus08hdQbobTLNoCXlo,5569
78
+ sglang/srt/disaggregation/decode.py,sha256=lqpmpQYMXzDCM19R4Pe4RF8q92UvBqKUITauFGDylEs,33335
79
+ sglang/srt/disaggregation/decode_schedule_batch_mixin.py,sha256=8UdNaj8KKMLV5Cydhw8gnHM-zRrnKM_AAd3Qc-SRfWg,5648
78
80
  sglang/srt/disaggregation/kv_events.py,sha256=yFbtwOeblcCmOXTjg00TopxnyjkmCBQIVz46KB8jetY,13555
79
81
  sglang/srt/disaggregation/launch_lb.py,sha256=mcbAztN4gnHevw_T5R2_nWsymsDEY9vHkm7OJ1vr6cc,4211
80
- sglang/srt/disaggregation/mini_lb.py,sha256=WGEdKQ0jMyRtZ1gG6q8uPnP9ovXkWGQamWZ3YmS4ng4,13429
81
- sglang/srt/disaggregation/prefill.py,sha256=xj_Q5NT0DP-5bxV0Hct6PnevAVbG0ZA5r3dTcT1ivRo,23102
82
- sglang/srt/disaggregation/utils.py,sha256=ih7KiJxyZF9Ma7O09UpBqkm8OCgfOvlg4odtoCK4uy4,10312
82
+ sglang/srt/disaggregation/mini_lb.py,sha256=BBeIdeZZxi7Ra5_hukHHX3TNdyfZ4tP1GBUxC-qrv_g,14401
83
+ sglang/srt/disaggregation/prefill.py,sha256=jWOqYXBBiNuOC78a028FN6mlklzEki4MjRdTA6UE5zU,23518
84
+ sglang/srt/disaggregation/utils.py,sha256=LBiRVbJ4jjYJXn0JL4VMTmGwAMAIGqJ_zaqPLjrmfkU,11339
83
85
  sglang/srt/disaggregation/base/__init__.py,sha256=4VwUv0aWxwmVL1049XK82aLTNxmt0WY5RPy9li-wyVk,160
84
- sglang/srt/disaggregation/base/conn.py,sha256=6KK_7HL7xixVSB8iap8jjJrIv3JgHDmynZ39Os3k5MU,2771
86
+ sglang/srt/disaggregation/base/conn.py,sha256=CPDAoAkYaFtVPLa1QROfwipSVe7MH6omzIBHzo8TSYk,2811
85
87
  sglang/srt/disaggregation/common/__init__.py,sha256=7yl-EGLMVKRpBUaGF_7lwAsw2J_mqpRZV0238VGxD9o,126
86
88
  sglang/srt/disaggregation/common/conn.py,sha256=CZR1lMCE_mpSkvjc6BBmSr1SbHY1uZuxjpanazD-YXc,16071
87
89
  sglang/srt/disaggregation/common/utils.py,sha256=SxRhAWisNK8seGhb5BXBJ5u53DF7yeKVPMWPcB5ywbE,1194
88
90
  sglang/srt/disaggregation/fake/__init__.py,sha256=jJGWdXwaQiGIoR6atKqkQfkJmVyQ09l55VUN2WjwaeY,77
89
91
  sglang/srt/disaggregation/fake/conn.py,sha256=oD1DArn1yDFZCu-X6p93uSLlAXEkt9lYxERICMznxGw,2286
90
92
  sglang/srt/disaggregation/mooncake/__init__.py,sha256=0TgqkAdQI1YynbHY6c0QISvVoOSk-0SwCIq5rjPSmgE,156
91
- sglang/srt/disaggregation/mooncake/conn.py,sha256=udV5-9Bg1CzBXAQasCYxdcxHmwTCGss4CHmsn7Vb_Rs,47871
92
- sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=gOcjZPtksJDq4iiIs0_zREdOM8trqUcmOXkrj-wiCuU,3149
93
+ sglang/srt/disaggregation/mooncake/conn.py,sha256=MATJkiS_5Vh8fc8Gx6S6cI3zCYDbgKXVEAa1xkUsCKA,58879
94
+ sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=JYB9T-EPdJNfv4I_sVpmMOZCOJ14itD97ws6tTvj240,4281
93
95
  sglang/srt/disaggregation/nixl/__init__.py,sha256=qODVPIGWUXKXq4zsRIcMYoAoAeg6nBIN9vdQOlVMANE,136
94
- sglang/srt/disaggregation/nixl/conn.py,sha256=KzVPRH8MaOAiG9EdfDN31sE9U9Ibsj2McDlSRWzHiP0,18055
96
+ sglang/srt/disaggregation/nixl/conn.py,sha256=G2l-FuXUvtsEo3Z24vyQ8iTcFjqG-sise4ItAtiny30,20327
95
97
  sglang/srt/distributed/__init__.py,sha256=jFOcyt-wFAPMBUAf9zkZalNQlt-4rqmT6pCKBz1E4qo,149
96
98
  sglang/srt/distributed/communication_op.py,sha256=IBnFUdMftK_VSTMMMitGveonorFUUVNL4guqO31cMSc,1130
97
- sglang/srt/distributed/parallel_state.py,sha256=l4vrGXIr9ykd0imtzKdMjnBrQPRvbrYg0jQ9trDkdzI,53527
99
+ sglang/srt/distributed/parallel_state.py,sha256=0_G1TtBOFMYDix5rfuEHYBMpy9A-OuPs9yFd5nCiZ8Q,53927
98
100
  sglang/srt/distributed/utils.py,sha256=aaCxATncLGnVgB0WlGpBdee0behKW8Dy_dakqcuKSaQ,8497
99
101
  sglang/srt/distributed/device_communicators/cuda_wrapper.py,sha256=3jvPG-Ow5UBLiXhfx8T8snR7crSZbPpARAggsDPWq7k,7038
100
102
  sglang/srt/distributed/device_communicators/custom_all_reduce.py,sha256=qje-PQ3v8yaV-oYVLPws1mgIlXVsGKFCOvXHmSe8ZXg,19624
@@ -107,51 +109,60 @@ sglang/srt/distributed/device_communicators/pynccl_wrapper.py,sha256=LblisImY9d6
107
109
  sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=d8mykYmXM1lfbPm8GNtqCF0Un_pdXYjbNmsgoVFyyow,20874
108
110
  sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
109
111
  sglang/srt/entrypoints/EngineBase.py,sha256=FCwqQMJE_8CCCMThAXUZaKafsHvh2BaQ31J-7_ormwA,2310
110
- sglang/srt/entrypoints/engine.py,sha256=SkBxdo6UDcQBOHbfpcfGrXsv0To92j_3BsdkA3Z6y1Y,30316
111
- sglang/srt/entrypoints/http_server.py,sha256=F-yBo8nKYhFqi-WYvrHOGLmWpghEUqsc__BzB7xjlcI,31965
112
- sglang/srt/entrypoints/http_server_engine.py,sha256=0QJFmrKAVnBmFbObHrmvUDcxDAsvnv6H29cQq286CKc,4939
113
- sglang/srt/entrypoints/verl_engine.py,sha256=RYizNetTHzcB8dErX1EW4NsyRNGkFPljYaAf7pVRPdM,7002
112
+ sglang/srt/entrypoints/engine.py,sha256=uSlN1vorCBCYk8n8AnCpbnCu-4p9JjukLyXJZZzTrGU,30283
113
+ sglang/srt/entrypoints/http_server.py,sha256=Fyb3z9OKXF9h_-duhWsHLjZayE3uoZ2dHpeV757bxXc,34745
114
+ sglang/srt/entrypoints/http_server_engine.py,sha256=ncN45ti9mawSOimPSedI6zugfoMhMQOYh4tmdfC9LcE,4936
115
+ sglang/srt/entrypoints/openai/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
116
+ sglang/srt/entrypoints/openai/protocol.py,sha256=QqC3XeJ0RxlFBsPuobjtf6c7iRkEOC8Td4E_GvjKwxM,18759
117
+ sglang/srt/entrypoints/openai/serving_base.py,sha256=5NJ2S_6B2NFSwn4nLp6eaeJ5iC3IcQzMEY9lW_gPcdA,5246
118
+ sglang/srt/entrypoints/openai/serving_chat.py,sha256=tWKWjspTpNnuk-Aqfl_S6OmpZe6gz5XxmSGEwfIvCsY,35935
119
+ sglang/srt/entrypoints/openai/serving_completions.py,sha256=emIutVmnJgOgGRuAqdB80-kVHUfQbE67n1VS-76o4QY,16690
120
+ sglang/srt/entrypoints/openai/serving_embedding.py,sha256=hqPt2ELpT1yQ8sBXo801aLJ3sExoPXs_K5ZQtC2vpAs,6274
121
+ sglang/srt/entrypoints/openai/serving_rerank.py,sha256=8n6y5kC6UhRpWrDtzH0eIp0dznW12FT60MLCR4bvD2U,3333
122
+ sglang/srt/entrypoints/openai/serving_score.py,sha256=ebAts-m6Pq-LTgFKwggkywBUrAgUSppHHVEzgwYHUzo,1955
123
+ sglang/srt/entrypoints/openai/usage_processor.py,sha256=9LTB5rqdRuMKyZrIXiUBuF_WKaSg9X45YdzERDxbtCY,2746
124
+ sglang/srt/entrypoints/openai/utils.py,sha256=DT9quW0F1DX1P9NAuMxpERgXcNYKKtm7rd7J1m7EaDQ,2112
114
125
  sglang/srt/eplb_simulator/__init__.py,sha256=HIZaXqsvsOCMMJ81dSs1afSGZ5G6OiOZEtHpltyzzjY,21
115
126
  sglang/srt/eplb_simulator/reader.py,sha256=R9wV7PqaCjILSnfYk90hNY95iJeST08a55Uj2Qk4iE0,1832
116
- sglang/srt/function_call/base_format_detector.py,sha256=Ol8VaVrilkBhnjUUmWlt6E0eazvFd5gFrzNx0_x0_iw,12755
127
+ sglang/srt/function_call/base_format_detector.py,sha256=eMSqbx99YPm2m_ChLafFRuLOzXBtA0oJn_Q4FtB3FnM,12950
117
128
  sglang/srt/function_call/core_types.py,sha256=wLUsW8mUA-i-ISz5QUDL7Ejq72W-K1HIhFa9Wm-_oA8,786
118
- sglang/srt/function_call/deepseekv3_detector.py,sha256=FO5wbBMj1l9culkmZeC-LKguyRuqHcQOyX7NCEgfzTQ,8962
119
- sglang/srt/function_call/ebnf_composer.py,sha256=02oDBr_dhxqlpCFBLItodSPbul9xEHlet7O7KkWZEZo,10686
120
- sglang/srt/function_call/function_call_parser.py,sha256=8YqFlyxFHkp0DlUxe0TabTdV-Qx5SbaoURZrc8SfNJ8,7674
121
- sglang/srt/function_call/llama32_detector.py,sha256=_d_pdzhn6Ut0jsivurki797hPK3vcNYjmXUkJ5sPKbk,3611
122
- sglang/srt/function_call/mistral_detector.py,sha256=D2udBa9K1J6DdSPyJCp6JpaqYnrOMwoHYCjbHMWHxCc,4454
123
- sglang/srt/function_call/pythonic_detector.py,sha256=waFdfHJuNg9DanTRDNWgh-Ji8TQkebcldUDPj0Xm8kk,8723
124
- sglang/srt/function_call/qwen25_detector.py,sha256=YYDXHBcW1EQpZd0xzL92E6NBLG5t6vqPYOmFRACMdcw,4874
129
+ sglang/srt/function_call/deepseekv3_detector.py,sha256=XudIiIt2Mejokg6qOkX7vYS-3XIFstOBHv-m-Q5fBLQ,8970
130
+ sglang/srt/function_call/ebnf_composer.py,sha256=1r5UPEkUeMCug-P-c1mAZbJPgEXU8hR-fDLdYG5Q-wo,13585
131
+ sglang/srt/function_call/function_call_parser.py,sha256=IB3QXZjedOFhXcLIOWSiceE_dOqGPcHqqnALu3w-_VY,7682
132
+ sglang/srt/function_call/llama32_detector.py,sha256=EHYGAy1NCIxFWKdSQByPxyJgFaFRCtGdgj4ntEAgi00,3619
133
+ sglang/srt/function_call/mistral_detector.py,sha256=xNuVl2vDXVYbXyiXLkJZ9VM6njcwlS9EqryR1eJbMr4,4462
134
+ sglang/srt/function_call/pythonic_detector.py,sha256=rtXSflE4w993a4OqphWG-WQzvwRz0v3dfgHOnte1fpI,8731
135
+ sglang/srt/function_call/qwen25_detector.py,sha256=9JfZem_5nw91Og2biwq8eIpUQjy_3kFz4TQI8Lc3Vow,4882
125
136
  sglang/srt/function_call/utils.py,sha256=__ImDF2kNyoLWsYO5RYoryvy1mmgEjnjXlCvLv-uLCM,1695
126
- sglang/srt/layers/activation.py,sha256=GNp9qww-EF6JCvTY5_FVWtqodr2j6m0lhP7vmksnBNo,6647
127
- sglang/srt/layers/communicator.py,sha256=_H-ZkyDeu8Y1v4BkeiexHKsx2MK3qQCPFodoMqxN80U,18639
128
- sglang/srt/layers/dp_attention.py,sha256=oQ8qMzcv3r-x0GW8_j_D-BaGIM5BhQ_ehi-TkqIE4O4,9274
137
+ sglang/srt/layers/activation.py,sha256=w8gr84LdpbBxmSjK9cfqFGXiGxALOwkBzqHChdZ6z7M,7327
138
+ sglang/srt/layers/communicator.py,sha256=WbefauUNbwfAtaBySi-rqqXkoFZZpxdOJURLBHpF5qA,18597
139
+ sglang/srt/layers/dp_attention.py,sha256=e-AgUTa70NsNgsw4hB1a-B_yDv7T8PyXW3jqR8sIKgY,9807
129
140
  sglang/srt/layers/elementwise.py,sha256=XCrR2i-9dP-H6jQo2zUuquwZrsl_wEQqj5Wxk6WUf7o,13987
130
- sglang/srt/layers/layernorm.py,sha256=VW46eTmFLFo7KezfwMKvwXlC_NPUVMGb93K2WREgldY,6325
131
- sglang/srt/layers/linear.py,sha256=xdAv5qJodLdapHi9ex3e0Qa_KHldcGwEiE-LpLRs1U8,51939
132
- sglang/srt/layers/logits_processor.py,sha256=KGYC4rPTLCHRnV-yL2E3ACj-bxO0BVZcQ_ZEUl9UNgk,25207
141
+ sglang/srt/layers/layernorm.py,sha256=x6VDTFxvcJMx2txpBW4Y7G1sWfKYsksNpf-L4_ySSDo,7660
142
+ sglang/srt/layers/linear.py,sha256=ToLkotx239ze3rwizk3r05Gg7_LJk39hAdjgqWwYusE,52462
143
+ sglang/srt/layers/logits_processor.py,sha256=TTR7LgSwthaH6Qfmcda2Ampibtt-JcwpLSs6-OJI_sQ,25604
133
144
  sglang/srt/layers/multimodal.py,sha256=YVR69WW-2aGDcZHT8IVJ6F_LRM7wraZr8VjrPDXqDmA,2104
134
145
  sglang/srt/layers/parameter.py,sha256=zqWyEzpWzP4NNTjq3G9khq6XofgpcmJqQLg6Vd4WyWE,15084
135
146
  sglang/srt/layers/pooler.py,sha256=uZ6WX1FLMEafZwusyZdm6KuVlIwSjbKrdwk2qzgqNGk,3812
136
147
  sglang/srt/layers/radix_attention.py,sha256=IlqRB4bk06FOH05_7zB8lik0xLpys7jFooLeCwdO0j8,3437
137
- sglang/srt/layers/rotary_embedding.py,sha256=G70hvJbPV0Jeh80wuEUCNc2AKB-P9sJIscPHoRvoFT0,48011
138
- sglang/srt/layers/sampler.py,sha256=h30VHnKyNVtTkRvOhQWIAFvCY1abx2fw1d5zeE8N57o,11169
148
+ sglang/srt/layers/rotary_embedding.py,sha256=iOxdJEw9jhxYih7WpNkzlGybzewNu9GgqU8dLM-TyD0,52209
149
+ sglang/srt/layers/sampler.py,sha256=xNds1migup2s6b9_pS6ljkJUkvNtv7nmTGeIdOzoQ6w,11182
139
150
  sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSahyMfA,4063
140
151
  sglang/srt/layers/utils.py,sha256=IWGg1Hb7c33Z3LHRPVJyUAzp3BnSid23ZWXAmJ_Jvp8,1204
141
- sglang/srt/layers/vocab_parallel_embedding.py,sha256=GPKCQiR2yH5Z2XSE9g3vWObdegcwPnMHXv7qEZovLH4,22755
142
- sglang/srt/layers/attention/aiter_backend.py,sha256=CXFcu-PwUjhhQj02AWQop3lUzsvAF3K5OsGjz0Qahqs,32797
143
- sglang/srt/layers/attention/base_attn_backend.py,sha256=HCdOMceZQAmIzZTewwp1_EIRfb1URlQZ3uGUmDKWw-o,3445
144
- sglang/srt/layers/attention/cutlass_mla_backend.py,sha256=cZzKjxN7pm0J75dDrKMNKSaMVbZZGncM9mXS9CwqWLI,9784
152
+ sglang/srt/layers/vocab_parallel_embedding.py,sha256=65N0e4PlOwoTRZC9QpC2G83Crn-OI4rY9wZTRnOVNvg,23166
153
+ sglang/srt/layers/attention/aiter_backend.py,sha256=7sEUgViw-xl3yok91yyOD9gTi8lQmME0g0ZiKVTCcyI,32851
154
+ sglang/srt/layers/attention/base_attn_backend.py,sha256=KXVcCguwXh-PSrY9Y2aUrlXXUhWdbVxqVEF2_xIMvm4,3466
155
+ sglang/srt/layers/attention/cutlass_mla_backend.py,sha256=SIR7sKCCegwzahSz82I3gsDyN5TkKoa4yG4-pBQWBi4,9813
145
156
  sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
146
- sglang/srt/layers/attention/flashattention_backend.py,sha256=9U2_gxJZafBoSPUM34sVtYxESHBHCnyZVr6eUHEjJWU,92296
147
- sglang/srt/layers/attention/flashinfer_backend.py,sha256=a2PsYEoNrmZ7aXG9FDcPebkaBgUQ5DOcf5AWtte7yi0,49567
148
- sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=I_94zj9BlbUPc9Do4E2WE75xUA05xB_QZWYOvWv6hvI,33957
149
- sglang/srt/layers/attention/flashmla_backend.py,sha256=spXbr5Dtl9u4saiBcxAFbsDktbidexQIpvX6J2ZffZA,20615
157
+ sglang/srt/layers/attention/flashattention_backend.py,sha256=j4rnX5IARna-pZmVnSR9kjf_yINGnkKO6itTLwkx82k,93278
158
+ sglang/srt/layers/attention/flashinfer_backend.py,sha256=Ug4SkZnuHjBBwPOj2TfLlg0eU_GoZvKjhY4oYRU_qqU,49666
159
+ sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=NirRlNFxD0S2EncvvmrerIxNfuTnYpvemZV3MR5_FnU,34040
160
+ sglang/srt/layers/attention/flashmla_backend.py,sha256=5iSic5ho-lkXNas9mR3uLbXbEl-do31gc8gjR7-a79k,20711
150
161
  sglang/srt/layers/attention/intel_amx_backend.py,sha256=hH9lxXs_o66_IVs58oPhRCOXGUrHbXSz82FKtHr9GO8,4092
151
162
  sglang/srt/layers/attention/merge_state.py,sha256=OnWmNV6LYTUj0y8SCy_IK7A9q6v9bZwR4_XiLOIvbrk,1423
152
- sglang/srt/layers/attention/tbo_backend.py,sha256=5EuuVA3MEeooQ5jkvANYODZc92lVlTRuijo0ZELixDw,8092
163
+ sglang/srt/layers/attention/tbo_backend.py,sha256=xpU8pR_xOwhfWnmAgLL892nb0ujBQWxIbHZAL_d84zI,8175
153
164
  sglang/srt/layers/attention/torch_native_backend.py,sha256=RsI4lYiQJAj-Xy0vKkzbyKOWHAVPOD9b0n0dEehwlos,9496
154
- sglang/srt/layers/attention/triton_backend.py,sha256=Ao6hoiknuOKuXT6kGltMHbYEx1dCM00C3EzbwvyyPSk,36644
165
+ sglang/srt/layers/attention/triton_backend.py,sha256=AryrPc4kzioxkiucPkApMA__0HoQTKOHNpwN0GEgm98,36923
155
166
  sglang/srt/layers/attention/utils.py,sha256=KiRSIHaGBhErUhPWCSD3Uiaqb0pSEF0NJFwTTQXoMks,2887
156
167
  sglang/srt/layers/attention/vision.py,sha256=-wGBZs9TSY8px-KidPQpEevZiQd577HJ-Wl5cTKQgWM,16291
157
168
  sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=eN0Lqpu5h27lMPuRnisX4jrU0WR061pCFSrh3vLlSD0,18936
@@ -162,16 +173,16 @@ sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=waZsmpKIp8rTg
162
173
  sglang/srt/layers/attention/triton_ops/rocm_mla_decode_rope.py,sha256=664WnAJ91EiCUZOcnVDfbTQf4uGJ4ZDZB1CbxpEUFZc,13866
163
174
  sglang/srt/layers/moe/cutlass_moe.py,sha256=--bNTA2BGbHeULb_XXDoRUyWWbE-doHo5K5k1T0N0WA,14323
164
175
  sglang/srt/layers/moe/cutlass_moe_params.py,sha256=9NRCmgP_Ug3gGqCcpi-x-QRbLjCNpw8792gKXwZsbEU,6522
165
- sglang/srt/layers/moe/fused_moe_native.py,sha256=FIES6H7oS-XjwVsWSrAkQzTdnp7kq-Z5VoViqOHpmtg,4851
176
+ sglang/srt/layers/moe/fused_moe_native.py,sha256=bW3KWxxz9rxKMUQqfmAtF-7ptTODA1pwLydE05ABDJE,5030
166
177
  sglang/srt/layers/moe/router.py,sha256=5Aeqoix_AS4uymb665OJE904wVSBkQeFdZP4e7KKPvg,10530
167
- sglang/srt/layers/moe/topk.py,sha256=7l3oouJSdVfwT0KyZce74EKk5y7otFjGhGmL8IQWIX4,15216
178
+ sglang/srt/layers/moe/topk.py,sha256=_hIyTURqx6Id1C3NQmHVuPxivMN0ywmDoyhFvtgHIZY,18624
168
179
  sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
169
- sglang/srt/layers/moe/ep_moe/kernels.py,sha256=EhCYANB_y1_NpW-mYDMw0jYCzftlVNLcm4fqbB4xK7I,34975
170
- sglang/srt/layers/moe/ep_moe/layer.py,sha256=sXwTcjWeC1Ka_UW3Jzu7tG4PXXIEZvKiOEEdzrqt190,45006
171
- sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=wPc0ewfBeoyYmgKMbp1YoT81OaJSme7AmftrXbxR7Jg,23523
180
+ sglang/srt/layers/moe/ep_moe/kernels.py,sha256=xdYak2dkrUJjmUigKJ-GbWfdf-tXlUKbvtJgxVekbMA,40130
181
+ sglang/srt/layers/moe/ep_moe/layer.py,sha256=U-R2mffzW_snuAxvX1AfN0xgcbt-w9fNFUidd2Hi4eQ,55783
182
+ sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=IR5RtdL9aIG04QaeySsq4Oy-S8obivBBZKbFfPCGpXc,24077
172
183
  sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=h9yMFAL_bagUf-qBED8gSWdCOb7d8IdA-pE-L_nIg8E,842
173
- sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=sIv67ShhS-6QeZxncNN_R_sH_SbfL2uW7yUAs8xolN0,62952
174
- sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=ptJveWp0alXFiwvL6FyXOH5Us4YPbKj4bg5om8JuEmE,27073
184
+ sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=gM_nctg15Wc06pxiMq0IRE0QBCfNyebSsWwm5zdM_Uk,63225
185
+ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=AAEb9pvd3bxiuvIKV2FZboWNvffccYmhF9R09SQSRlo,32038
175
186
  "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
176
187
  "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
177
188
  "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
@@ -331,28 +342,29 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=ptJveWp0alXFiwvL6FyXOH5Us
331
342
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=272,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=Piw4LN6d8QYrUahWsw3XUOtTMD1o3vHPwA94sGI56Gk,3242
332
343
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=288,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=3T8_rF2PEojhgTMyQ8DscXgJCWWdWfDPj4M434zWcA4,3243
333
344
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=tme0ydWzIxdABZLk4tU8G_X2dJUYGGZNkQzNGcmcvUc,3261
345
+ "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=384,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=LNx1W_BsWZcpHomiScCRap46dV0-F7S_w3Htskoqlm8,3263
334
346
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=fo1akCuGoqcGwQgfh56hApgg-wLXfo9kHHksE_6m1F4,3262
335
347
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=a9XJZ0XudUTebAdN5guVq-3XSvZetHIRjatO1-_0xCk,3257
336
348
  sglang/srt/layers/quantization/__init__.py,sha256=isMl8LINJFMzEse4r2Ta_7JXT15eKCX4mDIikuwticM,12559
337
349
  sglang/srt/layers/quantization/awq.py,sha256=KemDG55U3B6YZVjMV71awVAIj0islFvtxcUHmOBeGy0,6739
338
350
  sglang/srt/layers/quantization/base_config.py,sha256=jWk_egQrVNMYmQgbTI9vkcgzScLFjB5_sywFlAfE5J0,4776
339
351
  sglang/srt/layers/quantization/blockwise_int8.py,sha256=vWyPZsRLhdKtSmjvlT5fsowBK_nEebYbDAUh2yqseGw,15285
340
- sglang/srt/layers/quantization/fp8.py,sha256=Odr96VvYrnPhJR-bH7_JZlduqd_6ikihjZQ1EhoGuQ8,44474
352
+ sglang/srt/layers/quantization/fp8.py,sha256=Ne3K177lBdWI8TXyJs8qSaH67KaAv2j4LrZeHyqyH_8,44678
341
353
  sglang/srt/layers/quantization/fp8_kernel.py,sha256=pUda_glnAprnFIj3VUgCUYMKb2-uK3UOC3yPahgRMBQ,34743
342
- sglang/srt/layers/quantization/fp8_utils.py,sha256=_90Js9EwurcSKIr69_6avEXbX3--Nd9LicRkRU6CGts,25580
354
+ sglang/srt/layers/quantization/fp8_utils.py,sha256=QYHx_OLXFxFCmSMgoKmbJ3Vgl4mVEcXykdnhHO7tU0g,25650
343
355
  sglang/srt/layers/quantization/gptq.py,sha256=d1frUjvXmZfQKkcMQY5t0BA4sXWHE9Jze24qxniptJE,26719
344
356
  sglang/srt/layers/quantization/int8_kernel.py,sha256=GfRC9FOn9exNvK4QHbUeBj3Hhv32VcyGphapFPt5b84,12625
345
357
  sglang/srt/layers/quantization/int8_utils.py,sha256=YK9CS-lb_n91kNCTKK5o5apYF31V2giDg5G5VKrpcUA,2356
346
358
  sglang/srt/layers/quantization/kv_cache.py,sha256=_9pF5rwvB7ta6Gdc5YKVVGbNzYwqmhIx4TrX1-xnodQ,3261
347
- sglang/srt/layers/quantization/modelopt_quant.py,sha256=P8k89x2HeOCNWTqlHfHMIDL3g1Jv_32HiSEh2tblKh8,28744
359
+ sglang/srt/layers/quantization/modelopt_quant.py,sha256=wW90ZS1ZrGOXf7l5svudmXqAi3JE2SkCV-fwqWeYfdg,30905
348
360
  sglang/srt/layers/quantization/moe_wna16.py,sha256=PBgcQULe9oD2i96BeJRZtIdIUkWzS0IWU9ev3C-h7mw,19646
349
361
  sglang/srt/layers/quantization/qoq.py,sha256=adhsCixRUTh1m8sfpRWNhh5oX3I1WJli3Elg8uexP6A,8222
350
- sglang/srt/layers/quantization/utils.py,sha256=AXvGD8KRZVVrkRR1Y64fGkz4lkUP-CAjAQdp0LDNXrE,5266
362
+ sglang/srt/layers/quantization/utils.py,sha256=qHdWbLQD8teKhv2tOcuyGegpFonXygoJHwCAD7Ur5MA,5439
351
363
  sglang/srt/layers/quantization/w8a8_fp8.py,sha256=ZB6ydquyPOORDe9OCoEBRH6dsQhCeRWv1YIgm8UDQwQ,11622
352
364
  sglang/srt/layers/quantization/w8a8_int8.py,sha256=_BG6Zo4tis0UGZ47EVfBmfpOxRKzXvd3EM15qDWy7OM,9147
353
365
  sglang/srt/layers/quantization/compressed_tensors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
354
366
  sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=EaOKuIA0zXwqmH_eVhWeNdGJT9d1d9gVvFyYkgpdjDg,25665
355
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=HMON7sPz0kfG5f7cDbPtup50Xc8G_nIYzTB0Eyc8-gg,26460
367
+ sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=w19mhCnOre3j6L2-Tc_sRtBPgwBOAKV4QmYLpzyU_Fk,26637
356
368
  sglang/srt/layers/quantization/compressed_tensors/utils.py,sha256=mnUmKWFQUnY8bVoFHUuNVwqsfS-cefeR-ofyaihCXcY,7621
357
369
  sglang/srt/layers/quantization/compressed_tensors/schemes/__init__.py,sha256=HWMTnmrj-mUCRXgcOwnnXLrvrAE-ONdPTSzSImjHCMA,347
358
370
  sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_scheme.py,sha256=tdKJC8c3SX8T3z8JL-1YCsg4ftcv55Wxt0vZrYftpX8,1635
@@ -517,9 +529,9 @@ sglang/srt/layers/quantization/deep_gemm_wrapper/entrypoint.py,sha256=On_uvP1Thx
517
529
  sglang/srt/lora/layers.py,sha256=FaTYf0w3zbKKKAT1nJdAzGRtxvySB6FpNKXvEi5RHH0,12958
518
530
  sglang/srt/lora/lora.py,sha256=WY3cISWrQ5Ye3YcNKJUwAWRImzlpLQva2fGepg6P5EY,9744
519
531
  sglang/srt/lora/lora_config.py,sha256=qDgMTx_69jyJUl29O5FxLzYa0BMhqYVXWXfyyVOvGm0,1684
520
- sglang/srt/lora/lora_manager.py,sha256=8DG_TJQeHIGXLjSxJt7nFIaEQqgQKjdR4HwbokwDntI,13708
521
- sglang/srt/lora/mem_pool.py,sha256=lLmKinsTK3rWTNLJSLSSwKJud98z4up2Lsnb2b5H3ts,10007
522
- sglang/srt/lora/utils.py,sha256=3UwpyNONBW8X-ycgDV1lvVD3w_v1mVIs0T2wpVWVe7A,5484
532
+ sglang/srt/lora/lora_manager.py,sha256=pXQuS30X1KaqzDk1bD6Cb-cClcGopJcoNvGpXMshe54,18316
533
+ sglang/srt/lora/mem_pool.py,sha256=M4l4RO4tRjii5T4fq89nUoKJNHusaQHwvLrkryBCb1U,10132
534
+ sglang/srt/lora/utils.py,sha256=f3eySsA2tqlzMkvqwtRPWKIRjDiMk6ZFh9CJfUYdeqI,5486
523
535
  sglang/srt/lora/backend/base_backend.py,sha256=EIz8I-GIrdmK4fISw3ENhbJVVITaxKfyLxHXGPU4fPs,5044
524
536
  sglang/srt/lora/backend/flashinfer_backend.py,sha256=el6IAB4kTgDTbwCggmqFuukliyoapN5X6FLksG-4wJ8,4151
525
537
  sglang/srt/lora/backend/triton_backend.py,sha256=uqwBGlguXX8EkCKjSPqac6SPTgcKA31u3u6HkKPQcos,2530
@@ -528,33 +540,35 @@ sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=vOUDzC_Sq7UGnDbdF8F7ChKG12hD
528
540
  sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=qCpJAksCSXS3Znm-ABx1QfrdNqj9hrP61oaeW-n2RhU,5945
529
541
  sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=yNBgl8jS76DPlGJg8puRkSnKkewf2_2oKfVZZRq4lYY,4351
530
542
  sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=EoF9HymgQ1V9DwyOJOvbw1hCbEjprVOne_fC4VHgj-s,4518
531
- sglang/srt/managers/cache_controller.py,sha256=jv2-YkXRc032wnAoTQZMhTUzHrAjqSvdlYOSaWbbHr0,18483
543
+ sglang/srt/managers/cache_controller.py,sha256=K00t5dSZU6gbAMHZWhNyj0ndgHISvelTl7UMI5KqBuE,19352
532
544
  sglang/srt/managers/configure_logging.py,sha256=fOJaXAQ1n9m-8KPJndpsKvS885i69SMafoEADLIVfIM,1633
533
545
  sglang/srt/managers/data_parallel_controller.py,sha256=PZ-wOcAFn3PQqMB6I2vjIsFmplf0nlNl5hqTDKTHTG8,12112
534
546
  sglang/srt/managers/detokenizer_manager.py,sha256=SpLxTsSPKBZfD-ZMhJ5zpPPGuUb8PmcYgFSL9CsurU4,10696
535
547
  sglang/srt/managers/eplb_manager.py,sha256=YaxnvD1-wMV7BhtF9AxoNRVsJUdlQzFF7N-JDP43ojE,3411
536
- sglang/srt/managers/expert_distribution.py,sha256=HBcfNdS3l6ob17Z9KFlX5f79rqcaY41XMDp8emMD-TY,30880
548
+ sglang/srt/managers/expert_distribution.py,sha256=TzOSO7xFJ1VaxvbC2wqHq4l8UtWLX8K0rBmu5g_V2rU,31562
537
549
  sglang/srt/managers/expert_location.py,sha256=ZSsH17k5bAgbE1wuvpGaHGueiyhfPrgQakBFPu9jswo,16669
538
550
  sglang/srt/managers/expert_location_dispatch.py,sha256=U6-XLZ77RK0oy_JUVug2q-2LJjwoYX-js0_zhBNMXuM,4148
539
- sglang/srt/managers/io_struct.py,sha256=Bpxxn8ht8Q2T7jLEP5M8TK2_Vgn9XPWKvEz6V5yVX_s,33420
551
+ sglang/srt/managers/io_struct.py,sha256=gyP8JGyX4DSPYs4_0LIKUs4fj7tonrDhnreiSajgric,34022
540
552
  sglang/srt/managers/mm_utils.py,sha256=mA9W4xZBnXfs-4ZeALvhdpGAYxrJGfOxBWyoHVrt44Q,26518
541
553
  sglang/srt/managers/multimodal_processor.py,sha256=XlRYvNhF6XOssreRX9DZPhLSpps_VE62gSKw3EGdNPo,2088
542
- sglang/srt/managers/schedule_batch.py,sha256=2aKod4fNwHayeU9V5M9lFgiY2hliqoeWMPPSauo-Hpo,72438
543
- sglang/srt/managers/schedule_policy.py,sha256=6uRRgxbTJ0pztoAg8oRh2u1g6hNejDyMevD2-68zEYA,19723
544
- sglang/srt/managers/scheduler.py,sha256=Rq2mVboJha7ce0iGFG6xVjMIn-dyj75QMlIoUnuKf88,104546
554
+ sglang/srt/managers/schedule_batch.py,sha256=_Bt9hGWtiY3Dy0_GNijQVB9TT1b1V2xFOYK_M2aRvuo,73146
555
+ sglang/srt/managers/schedule_policy.py,sha256=0T8URzQmLvEmG-42-SFBBl9WnsOSwYO8-_CcBpuD38M,20474
556
+ sglang/srt/managers/scheduler.py,sha256=nejUktfqa3Qwf6TrN4CUVTYLELNwfUGFpExw-a8WFFw,110046
545
557
  sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=aaj0z1PD3fGIOPoTtWwqoKsii0yQCR7txVFXMPECveQ,29879
546
558
  sglang/srt/managers/session_controller.py,sha256=Lh1kruMcKqR7WVWYJRZbYgowtsssVlP7_paIVXLLIcE,5756
547
- sglang/srt/managers/tokenizer_manager.py,sha256=NcDNtH_7Bg2QE_v7YmNWDeLcaWFkEn4b-kO7i5KrTM4,65330
548
- sglang/srt/managers/tp_worker.py,sha256=cnI2JRxDqM7RXaaXAnLHNyYH4Jt7SB6HLAD5fRoWoNA,10025
549
- sglang/srt/managers/tp_worker_overlap_thread.py,sha256=W-FP7dkR9_voS0emMj4N5ML3Iao9gNZc-on6iUflbrw,9835
559
+ sglang/srt/managers/template_manager.py,sha256=RrwRA2oqId_PMQ98qJQGwIxMroOxiorl2sGC9ARou_0,8543
560
+ sglang/srt/managers/tokenizer_manager.py,sha256=zZR4c-jCm0n8LBn_5PT5eSv8Jptt50HM7OQeCvxBVqk,65586
561
+ sglang/srt/managers/tp_worker.py,sha256=A010h6fHS4gjsaqymQyAFiqLnZGUZiTi9rHIr3blmk0,10436
562
+ sglang/srt/managers/tp_worker_overlap_thread.py,sha256=jB9W7qSXKKnBrTep1yNKrYgLlylJvL-ab2Sa1ki6f0s,10349
550
563
  sglang/srt/managers/utils.py,sha256=9_VGE48EK0PXVJ26aYvbRJ6n7gIZALvCcf6uZCccCgM,1369
551
564
  sglang/srt/managers/eplb_algorithms/__init__.py,sha256=wVUv2ZhhC-_VbLaStpk3vulzhqExwfHGZJQqoohs-Fw,1963
552
565
  sglang/srt/managers/eplb_algorithms/deepseek.py,sha256=mMZT7zAAArccdRS0xXxifvMb3qn9enSt426uUTKeiq4,8340
553
566
  sglang/srt/managers/eplb_algorithms/deepseek_vec.py,sha256=Vzy5Iarua1VgVHzjBNZaVV_vt1LY1BLtJz7PmzB701k,10654
554
- sglang/srt/managers/multimodal_processors/base_processor.py,sha256=DQk5qNh7MDY9DQoASiQbCTwRb2HdvLWobzHmE0Ni_fg,21500
567
+ sglang/srt/managers/multimodal_processors/base_processor.py,sha256=1fl0eW24ju9FiKflieYEo2mooYFayWGVtamxhAtcWJM,23348
555
568
  sglang/srt/managers/multimodal_processors/clip.py,sha256=lRc2mcuDbAhZVf-0EfkO81pqDiol9zLvTpDqtPIBQ2k,1525
556
569
  sglang/srt/managers/multimodal_processors/deepseek_vl_v2.py,sha256=CMfhhdq7u6GzT8ZENo7ByClvQEx-HKaTGVgdYM1vMNw,3460
557
570
  sglang/srt/managers/multimodal_processors/gemma3.py,sha256=oBHXlbwto_84ZkjkW2A7F3Z7kNuDf039uDH4HVXKE1s,2290
571
+ sglang/srt/managers/multimodal_processors/gemma3n.py,sha256=UjAHeX4a2ZyPccCV_O9isxm61J-w5dglfhYO2IUQkyo,3578
558
572
  sglang/srt/managers/multimodal_processors/internvl.py,sha256=ASv3MQ0Ju6oZG7UceS5ziy4rL2d8Xf1_LbIFmEAuz2E,9512
559
573
  sglang/srt/managers/multimodal_processors/janus_pro.py,sha256=nSZYKLoCZtv7sQIM21KCt4jpnzVfcsF84m9CFPWwR7s,2058
560
574
  sglang/srt/managers/multimodal_processors/kimi_vl.py,sha256=8DER6QFDrmD0sZMjlAffY4z3jtBrrIYoU8ogpZIKNio,1868
@@ -566,25 +580,25 @@ sglang/srt/managers/multimodal_processors/phi4mm.py,sha256=-yTlt2kQkrgks2IzaRKmE
566
580
  sglang/srt/managers/multimodal_processors/pixtral.py,sha256=uQJYbxhafNCmg0xP52FdElnkUatAyintER1XNuKz3b8,4195
567
581
  sglang/srt/managers/multimodal_processors/qwen_vl.py,sha256=RLv8sBuKhAjeaVqrFR7hwEEw7fX1WOMZ6r2R4gFeV04,7025
568
582
  sglang/srt/managers/multimodal_processors/vila.py,sha256=AeQlhdIKKoN0lre8_zYQ7YpGu7I_rux1bRLt70sPKJ0,2594
569
- sglang/srt/mem_cache/base_prefix_cache.py,sha256=WOUIY6ATzltzXQ7QXS9biSY2o_kzbpDIuTD4C6TLCQQ,1019
570
- sglang/srt/mem_cache/chunk_cache.py,sha256=vcA-BI2MwlTiPHrzaufQKg-uYPvpeoizAI4bvv8Ie0w,2045
583
+ sglang/srt/mem_cache/allocator.py,sha256=gujYY5DFYQIz8VDAUOjATKtxUMpdRo6xkBunPF08SG4,12526
584
+ sglang/srt/mem_cache/base_prefix_cache.py,sha256=XHSzXKgBin-m1HsL47K-GobhLnajaGxqLqGtrLsWaZ0,2540
585
+ sglang/srt/mem_cache/chunk_cache.py,sha256=580NlAzr9Pm4yjvX5S_QP9ify998oOEPQuMpRye7uW8,2005
571
586
  sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
572
- sglang/srt/mem_cache/hiradix_cache.py,sha256=yw-vSz1BBMbjsw90k4i_XAwtlLm6ff-DL0bvEvTbJeg,16671
573
- sglang/srt/mem_cache/memory_pool.py,sha256=6qAHD-SLLYbOrJTIqGjgXBE577uwJrFvGknaiBHWzYI,26316
587
+ sglang/srt/mem_cache/hiradix_cache.py,sha256=Z_XgbLBxI8SN7kgsi1R6iX6AfgDUF5-ynVnBVRF7KUw,17061
588
+ sglang/srt/mem_cache/memory_pool.py,sha256=yLYDZ7-rLQi-nkhC_KuT4YmxnEOqM6Rq-j4_HAPehh4,27583
574
589
  sglang/srt/mem_cache/memory_pool_host.py,sha256=9N34Y7P0aM2xkmhsEw2_W8xN897h-HEwzRPIu1REwac,13186
575
590
  sglang/srt/mem_cache/multimodal_cache.py,sha256=Q-lYcI-3HoLu0WJhE2F_An8g9mkZ8LwMrYnQ4383Rf0,1266
576
- sglang/srt/mem_cache/paged_allocator.py,sha256=BrJS0vN1k-vTSgb_M8u_1KoZFRgzgR1WRyImCTq3T0U,9770
577
- sglang/srt/mem_cache/radix_cache.py,sha256=CeDEHnfYONX1Ra5QKemcs06RHzkDa8oV4ZaqHCayGI8,17699
591
+ sglang/srt/mem_cache/radix_cache.py,sha256=ojr9_bUwnPocmpbGZXz8JKac4dS-PrfNYk8UqF4Gvi8,17936
578
592
  sglang/srt/metrics/collector.py,sha256=C9QEJDOEdOPBwy2IJwFS3R6VbGzVzGs2xakKCCPvQDk,19903
579
593
  sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
580
- sglang/srt/model_executor/cuda_graph_runner.py,sha256=XoChArN7oF-YRApkZwiHHWRY8qFUpFPQ2o3k9K5zuJY,29279
594
+ sglang/srt/model_executor/cuda_graph_runner.py,sha256=MyZg9hw1dGcjIE5canm5TplUhFptIXt9FIKpDgXWjTQ,30450
581
595
  sglang/srt/model_executor/expert_location_updater.py,sha256=HWLY5lJAWefy2tobWJKlHs3qlBSCS57EwSMdfUuPFc4,20585
582
- sglang/srt/model_executor/forward_batch_info.py,sha256=zMT-foMHqR51bHZ2qj3KpWjJqDL5cfy7997mhiIFz1E,28848
583
- sglang/srt/model_executor/model_runner.py,sha256=1TyEuSOAwr4AMjtutCa2eXijcrK8Pe3-FlmXVcG9xG4,53496
596
+ sglang/srt/model_executor/forward_batch_info.py,sha256=ueHsjmGm52YqVK-8f-TRYpERCCdeSBuc2yaqxD9pWkQ,29268
597
+ sglang/srt/model_executor/model_runner.py,sha256=9EIHTDKRtXNrzURQYcC_rBjjWUPGnKHKYCV_GiW70zM,54490
584
598
  sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
585
- sglang/srt/model_loader/loader.py,sha256=JNhJrnCNCXTCroD5FKJdiBr8Dn4VNtTQ0XmqfkPI21k,56002
599
+ sglang/srt/model_loader/loader.py,sha256=-Pr8-YMaF3jIGXfhnEMM1WOloeWcPQnaUgSgWN6aAGI,57626
586
600
  sglang/srt/model_loader/utils.py,sha256=zSZBPA9ErPmkf-HfCxJjhmiFwYueB15KUg7NFspLvPY,4454
587
- sglang/srt/model_loader/weight_utils.py,sha256=yKnau-wH9muczoCpDTCVIqXFqz-QJmEEySplX3bMJWk,32153
601
+ sglang/srt/model_loader/weight_utils.py,sha256=jkd4R6wroef5A3xpVe6rst5xosVitxndnNwH6cMp_zo,35668
588
602
  sglang/srt/models/baichuan.py,sha256=HbvlErnkCSK4pRQYCSDxMcrn-1DQyfiNoeDcnRrJas8,15807
589
603
  sglang/srt/models/bert.py,sha256=ODJe8YfNRP-hHsomFWk4_QpcuiSsNfjzGf256EDS0Pc,15802
590
604
  sglang/srt/models/chatglm.py,sha256=cajLN9caBl09e0TwOFkiTTKDqwlbmHo_yS-NCjdeQW8,13957
@@ -593,8 +607,8 @@ sglang/srt/models/commandr.py,sha256=5Y_b3K0QY7D37nFGkyiGgY38RleRui_GJUYcHSuHUZo
593
607
  sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,15598
594
608
  sglang/srt/models/deepseek.py,sha256=ZnN02HdgXCB23Vno5V9UMUoOxH5HC82vNTwsVulUJ-o,17206
595
609
  sglang/srt/models/deepseek_janus_pro.py,sha256=OeeI7vZbE4HGpxa8CwT6-Lbfs7J7WMQ3oBNpVJQpv3w,70450
596
- sglang/srt/models/deepseek_nextn.py,sha256=2ziyjv_QEeT3lGNgxG_F-vcAT-K5p2hiz7VEimBgzLc,5675
597
- sglang/srt/models/deepseek_v2.py,sha256=TD9E-60eu7jlEqdW4nT1FQS5x03HzSVxOhhCdGc36nI,85363
610
+ sglang/srt/models/deepseek_nextn.py,sha256=FMeM-5oaUWhonnP7tP8oGvFympGoRkv8h9AfFocI-T4,5941
611
+ sglang/srt/models/deepseek_v2.py,sha256=KvRQntryvAu9IEY8OHrLhqIGND3qC-EKM5dr5sYhIXA,95084
598
612
  sglang/srt/models/deepseek_vl2.py,sha256=j8BdxZsMjm6lPdbDipEIKhVIVywCP1Vl1Kl46BZ5_0Y,13147
599
613
  sglang/srt/models/exaone.py,sha256=TpO-rtCpEZ8Ua7hGFnS8l2oAYhY0Pij50grc9WQ2mvc,13576
600
614
  sglang/srt/models/gemma.py,sha256=4cdrPISg1VKnsuI-QPTpYvet4BrX8BMKvCIN82iLskw,12641
@@ -602,10 +616,15 @@ sglang/srt/models/gemma2.py,sha256=kqtwdo93GWKm2iBN29RoIRH2ggRm-K_80LM5btgfBLo,1
602
616
  sglang/srt/models/gemma2_reward.py,sha256=V8U3_ADUHWPdOwvEe1jhGW-oJmBgL8t1TY3-67Ksv2A,2618
603
617
  sglang/srt/models/gemma3_causal.py,sha256=Vm605KeF7CBXbtxzOWF-v0TLbl0G12CLF-bEgTV9T0E,25197
604
618
  sglang/srt/models/gemma3_mm.py,sha256=b9YmkipsfVb5IXVeIVwW_PviXiCkRULhEsqNOvPoDxU,17221
619
+ sglang/srt/models/gemma3n_audio.py,sha256=isgKfjA5UieYawxU6medL2ssXlzYPqAbagDBnLcemC0,36405
620
+ sglang/srt/models/gemma3n_causal.py,sha256=nPGjcEOoLP-dhl7l94CB0XSC0g33ljFuIT_QeXb4BBE,36271
621
+ sglang/srt/models/gemma3n_mm.py,sha256=jFNhWCdPd4eChD0OlfSVtJfuufJr6qTj04c-oEXorQo,19273
622
+ sglang/srt/models/glm4.py,sha256=2VQzUqFkQTy_2nfkxP9SF6_9kKLTZUExGRjge7r99Es,11265
605
623
  sglang/srt/models/gpt2.py,sha256=kclhxEs8oJk1KCyhmAqo7rZqecVGGHYkc-a1WZi3aIk,9841
606
624
  sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtYhl4,10307
607
625
  sglang/srt/models/granite.py,sha256=5WOJyNYAlt5RNHSexNfPNihhSxIMd7wPzju1cTixKig,20852
608
626
  sglang/srt/models/grok.py,sha256=vESZeGS4adI_JAerXIkCcTm15-CNiGeS7VHc36C6w1A,28033
627
+ sglang/srt/models/hunyuan.py,sha256=dD9kWKTwh1DLa7b-laccQvh2PVVgAHx6487UT8VXhao,28994
609
628
  sglang/srt/models/idefics2.py,sha256=U3khd3hbdawJeRNXsxmaKHdssOCT5TPOZ1D-2_zHoQo,12079
610
629
  sglang/srt/models/internlm2.py,sha256=F_iNY1gUqzAjAuUatcE47gnrcoTh5_08PY2Rw9tKr9M,13150
611
630
  sglang/srt/models/internlm2_reward.py,sha256=ndfGmyqYZbVZ7C7rJ-v9oK3wa-EpoBGybS8MlyKZi2E,2522
@@ -622,7 +641,7 @@ sglang/srt/models/llama_reward.py,sha256=LF2nqMV5XOrljGjAwJg43mBv3z6Q040I2EYlgZe
622
641
  sglang/srt/models/llava.py,sha256=m0-qcEWEUyN7kIiCYJAZTi58MwL4ioqCAD99PSTpsuY,37496
623
642
  sglang/srt/models/llavavid.py,sha256=q0lHlRnoYHKJZsWnkIQdd6dYAQ26t7XsmrqA0zDGmZc,12829
624
643
  sglang/srt/models/mimo.py,sha256=Mp-iFp4YHuiuq-H8enUF5K5QbMnVcvEa6mURH6vM3yM,6140
625
- sglang/srt/models/mimo_mtp.py,sha256=y0wTNObGz0lyVt0zu7apdVlTRLEQmc_xsd8SvgSo_vo,7806
644
+ sglang/srt/models/mimo_mtp.py,sha256=jSmqJAXu7G3OO7jW1oa2suI4H_Yl7u5ZT7w4lHFbHhg,7292
626
645
  sglang/srt/models/minicpm.py,sha256=m5HFsSJj0Po09LY9R6qj6K4gceqWDMOePz3NDGgMGT4,14691
627
646
  sglang/srt/models/minicpm3.py,sha256=6-gfHSfXwyB5zw7AIj-c5TzjpEKR6CM0H67MrD8LVUE,19347
628
647
  sglang/srt/models/minicpmo.py,sha256=IBMFTmak4qokNl8zBXHLYfS5slPf88ECwox6CKHbiAU,76081
@@ -658,9 +677,6 @@ sglang/srt/models/vila.py,sha256=DRGJXP6iSv0XjgHYIJpMzBDNYZb3__Gg_wFkmZ5D66Y,986
658
677
  sglang/srt/models/xverse.py,sha256=DsNVI9JpzN4jj0Ry6aTrj7r-xq5YLOoDX2kH4YLJA-I,14035
659
678
  sglang/srt/models/xverse_moe.py,sha256=7KCM2-j12towDMNvXkuuYiBOmNauH6NG4Ip40x0khqA,16782
660
679
  sglang/srt/models/yivl.py,sha256=oToK7-u5IGO7xwpJIQ7VtudlK6-zPqJX4bt6_wv0SH8,4850
661
- sglang/srt/openai_api/adapter.py,sha256=OUXKtWrHOIsQly2KtecWo1gdTOzjCMGLHc3IXemgdjg,86874
662
- sglang/srt/openai_api/protocol.py,sha256=3PBC8kBhiLaLvbzr0yvuE0CIKCFgnxJQU62t9ADqhVc,16385
663
- sglang/srt/openai_api/utils.py,sha256=LL8VB-Qv9kNIVVWG1PYzoKMMHo7Y2NeVLMeJ9QEWVk8,6600
664
680
  sglang/srt/sampling/custom_logit_processor.py,sha256=AwODYVJdRkcQ8PGtJrhzKsqAgn8XZLQbAmR9fGiQzmc,1608
665
681
  sglang/srt/sampling/sampling_batch_info.py,sha256=JOGU-Fa8wPtytEJ30BHg1PldeNytwy8u31azLH436kw,13450
666
682
  sglang/srt/sampling/sampling_params.py,sha256=L2cU3MhAo9OvfdEtTdJidp9nTOkD6tiwlegxQ5vgIfU,6127
@@ -670,10 +686,10 @@ sglang/srt/sampling/penaltylib/min_new_tokens.py,sha256=rdU_D7RoIcrQPhysNQEzmr4T
670
686
  sglang/srt/sampling/penaltylib/orchestrator.py,sha256=XM-Lm6u7gYPtMZrTIc0FR4QxNZxBH5s_Cj82umyCzYk,5721
671
687
  sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=NRh10AJrrQlGJ6S-enGdRefrTrWpyqrSm-aNnyqQNQQ,2119
672
688
  sglang/srt/speculative/build_eagle_tree.py,sha256=lxfaTEOTY0N9QTlgXSUae3DVtPLqW8CcYXWcHsxCJp4,11632
673
- sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=D90K4mTVsHkDYT4izKoAbJoNUzIfSTq5K01TuG8i74o,9038
674
- sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py,sha256=qjEH9g0YynGakpfgznTXMm24vEhOROOpy-V6WNY7f60,9836
675
- sglang/srt/speculative/eagle_utils.py,sha256=UuyxfbBE9DNL9FPbzyW8Fqt9LLzMq85My5RIrlEny28,42793
676
- sglang/srt/speculative/eagle_worker.py,sha256=UOrT2Do9qvmP_gmXeJSzf8sP_sRLbZaBcIfr2cfUydI,35064
689
+ sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=fSNaHYJOQUgy2_AHRlZPYSQRIMgSiL_Pz5h53eothx4,14239
690
+ sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py,sha256=sdIHbcWwgsXDoo64-T5veDuYB5eHC9QBwu2Ns29SCSI,14930
691
+ sglang/srt/speculative/eagle_utils.py,sha256=cBQYqryCr4is5E_16xIcUue0Q2_Mn4meY8NCyuu4RAg,45558
692
+ sglang/srt/speculative/eagle_worker.py,sha256=fjYJmzz8vYCf3xod0dYoGmwyFWh6iTjnGfbV6SdY32M,38105
677
693
  sglang/srt/speculative/spec_info.py,sha256=rhaKG0TzyF9XZYHEWp1jccwTBohSNsUDvxHFtAoOl18,709
678
694
  sglang/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
679
695
  sglang/test/few_shot_gsm8k.py,sha256=qnEejCyPHGEMBEiNc7T4-t9lvoo2DNN72RXw-ei_TnY,4300
@@ -690,7 +706,7 @@ sglang/test/simple_eval_mmlu.py,sha256=FkwamjGMjueTixymkedF-YiPloSLiy4ftILFUrKZ9
690
706
  sglang/test/test_activation.py,sha256=GeTIJHxlLQfW3kM-X1FGa8Sa3dSGKHEXl5wEy-hfGis,1489
691
707
  sglang/test/test_block_fp8.py,sha256=1qBD-FJr5UaR9e5qDz91vDrS_DT1utnk82qeQ2bm-1E,21630
692
708
  sglang/test/test_block_fp8_deep_gemm_blackwell.py,sha256=Hnhq4kkyINHb4ONedkp5Kf7Xx0MjOoOBYL8J7ETvOVA,8180
693
- sglang/test/test_block_fp8_ep.py,sha256=TI0TVeoX8zRDtOFclonCfgSH6vnAIOGiMKQKpf9aKNs,10845
709
+ sglang/test/test_block_fp8_ep.py,sha256=Ufq6p32sb1RSbFKaSnsVF8RrqqwwAACBUgOJTEH2Lbc,10856
694
710
  sglang/test/test_custom_ops.py,sha256=2bSo9P5_rJZYFq8Y8IKRimDfFyZZGJluhL7Ngny0Pf4,5571
695
711
  sglang/test/test_cutlass_moe.py,sha256=beXdvpwpHh0O8fpLW-RV3Bx8Wu0lRFG_SNkL3Op5ITA,9905
696
712
  sglang/test/test_deepep_utils.py,sha256=749ysTBGNzh6rYUCJhhZBtZpeD15eWTeNHYCytcvZtc,7448
@@ -698,13 +714,13 @@ sglang/test/test_dynamic_grad_mode.py,sha256=L76yUCuk_ymNpXD2CmO8r2GiGjIvD_gtTsu
698
714
  sglang/test/test_fp4_moe.py,sha256=U4LHzpGl6pCiCZiENRPxVP6hxqkqRZ_0WXbB29vYeAo,8187
699
715
  sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg,3764
700
716
  sglang/test/test_programs.py,sha256=DJIbTChloPwaDiEWHQoSTuKv3BGx0-K8Y5e1mDDTACM,18910
701
- sglang/test/test_utils.py,sha256=xdi1-Fvs0ykmaYhYzC0Jx-n7nSFKSO9a6HAvWbXqVQo,36176
717
+ sglang/test/test_utils.py,sha256=UfVq6snSqzomW0sDTql54Vbefl6kVovbFx1R79y_mqw,36243
702
718
  sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
703
719
  sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
704
720
  sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
705
721
  sglang/test/attention/test_prefix_chunk_info.py,sha256=hpoDe2wfSa6RlUbfyri_c0iyBTb35UXGL9I2Xh6jamM,7772
706
- sglang-0.4.7.post1.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
707
- sglang-0.4.7.post1.dist-info/METADATA,sha256=DgH_mVdir5t4SV58Js1jZLSeJ5qB4bH0FW7aO3RfR20,26966
708
- sglang-0.4.7.post1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
709
- sglang-0.4.7.post1.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
710
- sglang-0.4.7.post1.dist-info/RECORD,,
722
+ sglang-0.4.8.post1.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
723
+ sglang-0.4.8.post1.dist-info/METADATA,sha256=isDKzDsTthshFCkEmL3isGMcgn1uBG3M2mvGolPE_xc,26609
724
+ sglang-0.4.8.post1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
725
+ sglang-0.4.8.post1.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
726
+ sglang-0.4.8.post1.dist-info/RECORD,,