sglang 0.4.7.post1__py3-none-any.whl → 0.4.8__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 (106) 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/constants.py +3 -0
  5. sglang/srt/conversation.py +13 -3
  6. sglang/srt/custom_op.py +5 -1
  7. sglang/srt/disaggregation/decode.py +22 -28
  8. sglang/srt/disaggregation/decode_schedule_batch_mixin.py +4 -3
  9. sglang/srt/disaggregation/mini_lb.py +34 -4
  10. sglang/srt/disaggregation/mooncake/conn.py +12 -16
  11. sglang/srt/disaggregation/prefill.py +17 -13
  12. sglang/srt/disaggregation/utils.py +46 -18
  13. sglang/srt/distributed/parallel_state.py +12 -4
  14. sglang/srt/entrypoints/engine.py +22 -28
  15. sglang/srt/entrypoints/http_server.py +149 -79
  16. sglang/srt/entrypoints/http_server_engine.py +0 -3
  17. sglang/srt/entrypoints/openai/__init__.py +0 -0
  18. sglang/srt/{openai_api → entrypoints/openai}/protocol.py +67 -29
  19. sglang/srt/entrypoints/openai/serving_base.py +149 -0
  20. sglang/srt/entrypoints/openai/serving_chat.py +921 -0
  21. sglang/srt/entrypoints/openai/serving_completions.py +424 -0
  22. sglang/srt/entrypoints/openai/serving_embedding.py +169 -0
  23. sglang/srt/entrypoints/openai/serving_rerank.py +102 -0
  24. sglang/srt/entrypoints/openai/serving_score.py +61 -0
  25. sglang/srt/entrypoints/openai/usage_processor.py +81 -0
  26. sglang/srt/entrypoints/openai/utils.py +72 -0
  27. sglang/srt/function_call/base_format_detector.py +7 -4
  28. sglang/srt/function_call/deepseekv3_detector.py +1 -1
  29. sglang/srt/function_call/ebnf_composer.py +64 -10
  30. sglang/srt/function_call/function_call_parser.py +6 -6
  31. sglang/srt/function_call/llama32_detector.py +1 -1
  32. sglang/srt/function_call/mistral_detector.py +1 -1
  33. sglang/srt/function_call/pythonic_detector.py +1 -1
  34. sglang/srt/function_call/qwen25_detector.py +1 -1
  35. sglang/srt/{openai_api/utils.py → jinja_template_utils.py} +6 -5
  36. sglang/srt/layers/activation.py +21 -3
  37. sglang/srt/layers/attention/aiter_backend.py +5 -2
  38. sglang/srt/layers/attention/base_attn_backend.py +1 -1
  39. sglang/srt/layers/attention/cutlass_mla_backend.py +1 -0
  40. sglang/srt/layers/attention/flashattention_backend.py +19 -9
  41. sglang/srt/layers/attention/flashinfer_backend.py +9 -6
  42. sglang/srt/layers/attention/flashinfer_mla_backend.py +7 -4
  43. sglang/srt/layers/attention/flashmla_backend.py +5 -2
  44. sglang/srt/layers/attention/tbo_backend.py +3 -3
  45. sglang/srt/layers/attention/triton_backend.py +19 -11
  46. sglang/srt/layers/communicator.py +5 -5
  47. sglang/srt/layers/dp_attention.py +11 -2
  48. sglang/srt/layers/layernorm.py +29 -2
  49. sglang/srt/layers/logits_processor.py +2 -2
  50. sglang/srt/layers/moe/ep_moe/kernels.py +159 -2
  51. sglang/srt/layers/moe/ep_moe/layer.py +207 -1
  52. 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
  53. sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +6 -0
  54. sglang/srt/layers/moe/fused_moe_triton/layer.py +75 -12
  55. sglang/srt/layers/moe/topk.py +91 -4
  56. sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py +6 -2
  57. sglang/srt/layers/quantization/fp8.py +25 -17
  58. sglang/srt/layers/quantization/modelopt_quant.py +62 -8
  59. sglang/srt/layers/quantization/utils.py +5 -2
  60. sglang/srt/layers/rotary_embedding.py +42 -2
  61. sglang/srt/layers/sampler.py +1 -1
  62. sglang/srt/lora/lora_manager.py +173 -74
  63. sglang/srt/lora/mem_pool.py +49 -45
  64. sglang/srt/lora/utils.py +1 -1
  65. sglang/srt/managers/cache_controller.py +33 -15
  66. sglang/srt/managers/io_struct.py +9 -12
  67. sglang/srt/managers/schedule_batch.py +40 -31
  68. sglang/srt/managers/schedule_policy.py +70 -56
  69. sglang/srt/managers/scheduler.py +147 -62
  70. sglang/srt/managers/template_manager.py +226 -0
  71. sglang/srt/managers/tokenizer_manager.py +11 -8
  72. sglang/srt/managers/tp_worker.py +12 -2
  73. sglang/srt/managers/tp_worker_overlap_thread.py +11 -0
  74. sglang/srt/mem_cache/{paged_allocator.py → allocator.py} +125 -34
  75. sglang/srt/mem_cache/base_prefix_cache.py +52 -8
  76. sglang/srt/mem_cache/chunk_cache.py +11 -16
  77. sglang/srt/mem_cache/hiradix_cache.py +34 -23
  78. sglang/srt/mem_cache/memory_pool.py +118 -114
  79. sglang/srt/mem_cache/radix_cache.py +20 -16
  80. sglang/srt/model_executor/cuda_graph_runner.py +76 -45
  81. sglang/srt/model_executor/forward_batch_info.py +18 -5
  82. sglang/srt/model_executor/model_runner.py +22 -6
  83. sglang/srt/model_loader/loader.py +8 -1
  84. sglang/srt/model_loader/weight_utils.py +11 -2
  85. sglang/srt/models/deepseek_nextn.py +29 -27
  86. sglang/srt/models/deepseek_v2.py +108 -26
  87. sglang/srt/models/glm4.py +312 -0
  88. sglang/srt/models/mimo_mtp.py +2 -18
  89. sglang/srt/reasoning_parser.py +21 -11
  90. sglang/srt/server_args.py +36 -8
  91. sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +131 -10
  92. sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py +125 -12
  93. sglang/srt/speculative/eagle_utils.py +80 -8
  94. sglang/srt/speculative/eagle_worker.py +124 -41
  95. sglang/srt/torch_memory_saver_adapter.py +19 -15
  96. sglang/srt/utils.py +177 -11
  97. sglang/test/test_block_fp8_ep.py +1 -0
  98. sglang/test/test_utils.py +1 -0
  99. sglang/version.py +1 -1
  100. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.dist-info}/METADATA +4 -10
  101. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.dist-info}/RECORD +104 -93
  102. sglang/srt/entrypoints/verl_engine.py +0 -179
  103. sglang/srt/openai_api/adapter.py +0 -2148
  104. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.dist-info}/WHEEL +0 -0
  105. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.dist-info}/licenses/LICENSE +0 -0
  106. {sglang-0.4.7.post1.dist-info → sglang-0.4.8.dist-info}/top_level.txt +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: sglang
3
- Version: 0.4.7.post1
3
+ Version: 0.4.8
4
4
  Summary: SGLang is yet another fast serving framework for large language models and vision language models.
5
5
  License: Apache License
6
6
  Version 2.0, January 2004
@@ -230,6 +230,7 @@ Requires-Dist: modelscope; extra == "runtime-common"
230
230
  Requires-Dist: msgspec; extra == "runtime-common"
231
231
  Requires-Dist: ninja; extra == "runtime-common"
232
232
  Requires-Dist: orjson; extra == "runtime-common"
233
+ Requires-Dist: outlines==0.1.11; extra == "runtime-common"
233
234
  Requires-Dist: packaging; extra == "runtime-common"
234
235
  Requires-Dist: partial_json_parser; extra == "runtime-common"
235
236
  Requires-Dist: pillow; extra == "runtime-common"
@@ -249,13 +250,12 @@ Requires-Dist: xgrammar==0.1.19; extra == "runtime-common"
249
250
  Provides-Extra: srt
250
251
  Requires-Dist: sglang[runtime_common]; extra == "srt"
251
252
  Requires-Dist: sgl-kernel==0.1.9; extra == "srt"
252
- Requires-Dist: flashinfer_python==0.2.6.post1; extra == "srt"
253
253
  Requires-Dist: torch==2.7.1; extra == "srt"
254
254
  Requires-Dist: torchaudio==2.7.1; extra == "srt"
255
255
  Requires-Dist: torchvision==0.22.1; extra == "srt"
256
256
  Requires-Dist: cuda-python; extra == "srt"
257
- Requires-Dist: outlines<=0.1.11,>=0.0.44; extra == "srt"
258
257
  Requires-Dist: einops; extra == "srt"
258
+ Requires-Dist: flashinfer_python==0.2.6.post1; extra == "srt"
259
259
  Provides-Extra: blackwell
260
260
  Requires-Dist: sglang[runtime_common]; extra == "blackwell"
261
261
  Requires-Dist: sgl-kernel; extra == "blackwell"
@@ -263,27 +263,21 @@ Requires-Dist: torch==2.7.1; extra == "blackwell"
263
263
  Requires-Dist: torchaudio==2.7.1; extra == "blackwell"
264
264
  Requires-Dist: torchvision==0.22.1; extra == "blackwell"
265
265
  Requires-Dist: cuda-python; extra == "blackwell"
266
- Requires-Dist: outlines<=0.1.11,>=0.0.44; extra == "blackwell"
267
266
  Requires-Dist: einops; extra == "blackwell"
268
267
  Requires-Dist: flashinfer_python==0.2.6.post1; extra == "blackwell"
269
268
  Provides-Extra: srt-hip
270
269
  Requires-Dist: sglang[runtime_common]; extra == "srt-hip"
271
270
  Requires-Dist: torch; extra == "srt-hip"
272
271
  Requires-Dist: vllm==0.6.7.dev2; extra == "srt-hip"
273
- Requires-Dist: outlines==0.1.11; extra == "srt-hip"
274
272
  Provides-Extra: srt-xpu
275
273
  Requires-Dist: sglang[runtime_common]; extra == "srt-xpu"
276
- Requires-Dist: outlines<=0.1.11,>=0.0.44; extra == "srt-xpu"
277
274
  Provides-Extra: srt-hpu
278
275
  Requires-Dist: sglang[runtime_common]; extra == "srt-hpu"
279
- Requires-Dist: outlines<=0.1.11,>=0.0.44; extra == "srt-hpu"
280
276
  Provides-Extra: srt-cpu
281
277
  Requires-Dist: sglang[runtime_common]; extra == "srt-cpu"
282
- Requires-Dist: outlines<=0.1.11,>=0.0.44; extra == "srt-cpu"
283
278
  Requires-Dist: einops; extra == "srt-cpu"
284
279
  Provides-Extra: srt-npu
285
280
  Requires-Dist: sglang[runtime_common]; extra == "srt-npu"
286
- Requires-Dist: outlines<=0.1.11,>=0.0.44; extra == "srt-npu"
287
281
  Provides-Extra: openai
288
282
  Requires-Dist: openai>=1.0; extra == "openai"
289
283
  Requires-Dist: tiktoken; extra == "openai"
@@ -292,7 +286,7 @@ Requires-Dist: anthropic>=0.20.0; extra == "anthropic"
292
286
  Provides-Extra: litellm
293
287
  Requires-Dist: litellm>=1.0.0; extra == "litellm"
294
288
  Provides-Extra: torch-memory-saver
295
- Requires-Dist: torch_memory_saver>=0.0.4; extra == "torch-memory-saver"
289
+ Requires-Dist: torch_memory_saver>=0.0.8; extra == "torch-memory-saver"
296
290
  Provides-Extra: decord
297
291
  Requires-Dist: decord; extra == "decord"
298
292
  Provides-Extra: test
@@ -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=40-PUZPRIakJU2yYWQcwTYvSJA6iewqiG8XylhxuAQk,22
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=9Ix1NRQAErjXen8s4Yobb1HKizln7VBdrZPK_QKONoY,37104
35
+ sglang/srt/custom_op.py,sha256=At-Nqm_noJyoxi928TLvIPzUgj3DM-St-g2mdx1CWEI,2912
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
44
+ sglang/srt/reasoning_parser.py,sha256=vf0kWBM4IXwbuzGBIOMdiXdn9gavqkHb1QIaTbkU7vc,6742
45
+ sglang/srt/server_args.py,sha256=HCjiKk1VujrWMR2SPiO2bzgdnySGZO34PaoSeCM-pEw,72873
46
+ sglang/srt/torch_memory_saver_adapter.py,sha256=K_eTx0UU84MHSTXI3iqYLdHV4IWtJMJ2FKdGFJR8v1E,2417
45
47
  sglang/srt/two_batch_overlap.py,sha256=Q6tobqItnlK5ad9mqqMMx0HJ-PAzW4DgBtgSQ7Ax6sM,21848
46
- sglang/srt/utils.py,sha256=Zl-xEOFgPyr91A9-q-0Y1PUobAmKvKURgSDR3EBGk4U,72516
48
+ sglang/srt/utils.py,sha256=kQWtZ0JUyaYDJNW-zNV0AU7Yfn9qGnRztPoSM5gC8sE,78080
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
@@ -73,13 +75,13 @@ 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=GLIaPpm3I3oykq3A41Ru4gdsLa9LwUADc3xqM2LuHSU,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=SQYLDpanJNc8sJHXQoPD6P48HGAEigxdf2CVal3Nl40,23419
84
+ sglang/srt/disaggregation/utils.py,sha256=84EF9I0lOK8bWqd8cbTTgVDzG6ecofp6WuSVqgCe1jc,11386
83
85
  sglang/srt/disaggregation/base/__init__.py,sha256=4VwUv0aWxwmVL1049XK82aLTNxmt0WY5RPy9li-wyVk,160
84
86
  sglang/srt/disaggregation/base/conn.py,sha256=6KK_7HL7xixVSB8iap8jjJrIv3JgHDmynZ39Os3k5MU,2771
85
87
  sglang/srt/disaggregation/common/__init__.py,sha256=7yl-EGLMVKRpBUaGF_7lwAsw2J_mqpRZV0238VGxD9o,126
@@ -88,13 +90,13 @@ sglang/srt/disaggregation/common/utils.py,sha256=SxRhAWisNK8seGhb5BXBJ5u53DF7yeK
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
93
+ sglang/srt/disaggregation/mooncake/conn.py,sha256=ES9N4J_zy5xdG8_YEsrcodXZPCBwKf9SBddOBCGLOHc,47837
92
94
  sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=gOcjZPtksJDq4iiIs0_zREdOM8trqUcmOXkrj-wiCuU,3149
93
95
  sglang/srt/disaggregation/nixl/__init__.py,sha256=qODVPIGWUXKXq4zsRIcMYoAoAeg6nBIN9vdQOlVMANE,136
94
96
  sglang/srt/disaggregation/nixl/conn.py,sha256=KzVPRH8MaOAiG9EdfDN31sE9U9Ibsj2McDlSRWzHiP0,18055
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=oNcZDupOYllUjtYnNZOuOqfQzxg3SonwC4jNn-ByvQY,30169
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=JepHGUD8DcwNJQ0B2YnR6ZvWp6fhc4IQa2zQthBI8z4,17110
117
+ sglang/srt/entrypoints/openai/serving_base.py,sha256=5NJ2S_6B2NFSwn4nLp6eaeJ5iC3IcQzMEY9lW_gPcdA,5246
118
+ sglang/srt/entrypoints/openai/serving_chat.py,sha256=ojMNf55Nq-WVpwhl7TQdwepyb1JShfKJkzgfWCPH1UE,36542
119
+ sglang/srt/entrypoints/openai/serving_completions.py,sha256=7JKbykpFYnqsmej6iK_SHnD67IJ7KOVBbl11TVtKPPg,16661
120
+ sglang/srt/entrypoints/openai/serving_embedding.py,sha256=ZAX4qQ9YZ7qUvT6zCAfrGFtkEfydPHiWuoPDz2Qrr2I,6245
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=vJUWe5O72c0yX9jxb0TND_6SFhiOnvHtp9H67ftgYnU,7174
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
141
+ sglang/srt/layers/layernorm.py,sha256=OLjIEv9POyTOOj2G7rSNzIJ6kdA9DM4-azFZALyQhbw,7149
131
142
  sglang/srt/layers/linear.py,sha256=xdAv5qJodLdapHi9ex3e0Qa_KHldcGwEiE-LpLRs1U8,51939
132
- sglang/srt/layers/logits_processor.py,sha256=KGYC4rPTLCHRnV-yL2E3ACj-bxO0BVZcQ_ZEUl9UNgk,25207
143
+ sglang/srt/layers/logits_processor.py,sha256=68kkHIwW7mTMSBjUpuw8blLWdL6i3XtHeOcqt4ErYbY,25195
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=D5oj1CwmZ7Dq8I1j0hfKXcd2jn7hzu8Z5eMI7phUDWw,49469
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
152
  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
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=JgAp-douu45GYrkxc-d9LHE-rw7pXbVpishHfsU9m_8,92529
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
@@ -164,14 +175,14 @@ sglang/srt/layers/moe/cutlass_moe.py,sha256=--bNTA2BGbHeULb_XXDoRUyWWbE-doHo5K5k
164
175
  sglang/srt/layers/moe/cutlass_moe_params.py,sha256=9NRCmgP_Ug3gGqCcpi-x-QRbLjCNpw8792gKXwZsbEU,6522
165
176
  sglang/srt/layers/moe/fused_moe_native.py,sha256=FIES6H7oS-XjwVsWSrAkQzTdnp7kq-Z5VoViqOHpmtg,4851
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=fLjl0DZMn6UnUtbx8AjyE0wtAtjn0W64t1DrDXBbjiM,17548
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
180
+ sglang/srt/layers/moe/ep_moe/kernels.py,sha256=xdYak2dkrUJjmUigKJ-GbWfdf-tXlUKbvtJgxVekbMA,40130
181
+ sglang/srt/layers/moe/ep_moe/layer.py,sha256=Tl45TYm200db9JNJMuCxgX9yJpKPQ8hUioNkAhUsd7M,52875
171
182
  sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=wPc0ewfBeoyYmgKMbp1YoT81OaJSme7AmftrXbxR7Jg,23523
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=FWUfurqSu4NZBzA7oWx2fcqQGtAhPy-U5qgLsaB3lrI,63103
185
+ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=nWmrt8Ihybl3uo0tsDNt0IpyvoBx8QO6KRS4rrP8YL8,29833
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
354
  sglang/srt/layers/quantization/fp8_utils.py,sha256=_90Js9EwurcSKIr69_6avEXbX3--Nd9LicRkRU6CGts,25580
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
@@ -528,7 +540,7 @@ 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
@@ -536,17 +548,18 @@ sglang/srt/managers/eplb_manager.py,sha256=YaxnvD1-wMV7BhtF9AxoNRVsJUdlQzFF7N-JD
536
548
  sglang/srt/managers/expert_distribution.py,sha256=HBcfNdS3l6ob17Z9KFlX5f79rqcaY41XMDp8emMD-TY,30880
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=bWybydmTbTuKW_AO6VjHPhB5jbWFdA7qcxb5Ee7Txkg,33687
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=meRGsHztVoSWNHuyq-Qzb_0Xk88d4OasGTvTYBvKoI8,72781
555
+ sglang/srt/managers/schedule_policy.py,sha256=0T8URzQmLvEmG-42-SFBBl9WnsOSwYO8-_CcBpuD38M,20474
556
+ sglang/srt/managers/scheduler.py,sha256=FcnzE0JuheAvuZlslbZRF7sQIbRIcwMkKKHp6--mGxs,108539
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
@@ -566,25 +579,25 @@ sglang/srt/managers/multimodal_processors/phi4mm.py,sha256=-yTlt2kQkrgks2IzaRKmE
566
579
  sglang/srt/managers/multimodal_processors/pixtral.py,sha256=uQJYbxhafNCmg0xP52FdElnkUatAyintER1XNuKz3b8,4195
567
580
  sglang/srt/managers/multimodal_processors/qwen_vl.py,sha256=RLv8sBuKhAjeaVqrFR7hwEEw7fX1WOMZ6r2R4gFeV04,7025
568
581
  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
582
+ sglang/srt/mem_cache/allocator.py,sha256=gujYY5DFYQIz8VDAUOjATKtxUMpdRo6xkBunPF08SG4,12526
583
+ sglang/srt/mem_cache/base_prefix_cache.py,sha256=XHSzXKgBin-m1HsL47K-GobhLnajaGxqLqGtrLsWaZ0,2540
584
+ sglang/srt/mem_cache/chunk_cache.py,sha256=580NlAzr9Pm4yjvX5S_QP9ify998oOEPQuMpRye7uW8,2005
571
585
  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
586
+ sglang/srt/mem_cache/hiradix_cache.py,sha256=Z_XgbLBxI8SN7kgsi1R6iX6AfgDUF5-ynVnBVRF7KUw,17061
587
+ sglang/srt/mem_cache/memory_pool.py,sha256=yLYDZ7-rLQi-nkhC_KuT4YmxnEOqM6Rq-j4_HAPehh4,27583
574
588
  sglang/srt/mem_cache/memory_pool_host.py,sha256=9N34Y7P0aM2xkmhsEw2_W8xN897h-HEwzRPIu1REwac,13186
575
589
  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
590
+ sglang/srt/mem_cache/radix_cache.py,sha256=ojr9_bUwnPocmpbGZXz8JKac4dS-PrfNYk8UqF4Gvi8,17936
578
591
  sglang/srt/metrics/collector.py,sha256=C9QEJDOEdOPBwy2IJwFS3R6VbGzVzGs2xakKCCPvQDk,19903
579
592
  sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
580
- sglang/srt/model_executor/cuda_graph_runner.py,sha256=XoChArN7oF-YRApkZwiHHWRY8qFUpFPQ2o3k9K5zuJY,29279
593
+ sglang/srt/model_executor/cuda_graph_runner.py,sha256=OqU5EYdJU6pfk3fMNPklnvVhlMaXKWWKguLkAV9cdbw,30444
581
594
  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
595
+ sglang/srt/model_executor/forward_batch_info.py,sha256=ueHsjmGm52YqVK-8f-TRYpERCCdeSBuc2yaqxD9pWkQ,29268
596
+ sglang/srt/model_executor/model_runner.py,sha256=mbYBcddyYxBuRcFmRawa17xdxySxdG-LwxmovH4BCLw,54309
584
597
  sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
585
- sglang/srt/model_loader/loader.py,sha256=JNhJrnCNCXTCroD5FKJdiBr8Dn4VNtTQ0XmqfkPI21k,56002
598
+ sglang/srt/model_loader/loader.py,sha256=R1-_cWKvrFx6zWWDYxKHOnfP7QtnVmMXi_zyS6seeCI,56286
586
599
  sglang/srt/model_loader/utils.py,sha256=zSZBPA9ErPmkf-HfCxJjhmiFwYueB15KUg7NFspLvPY,4454
587
- sglang/srt/model_loader/weight_utils.py,sha256=yKnau-wH9muczoCpDTCVIqXFqz-QJmEEySplX3bMJWk,32153
600
+ sglang/srt/model_loader/weight_utils.py,sha256=XwNVS5MLtNzW5fTS5UDK2CHUuSDMHKznI4Du8l4xoks,32562
588
601
  sglang/srt/models/baichuan.py,sha256=HbvlErnkCSK4pRQYCSDxMcrn-1DQyfiNoeDcnRrJas8,15807
589
602
  sglang/srt/models/bert.py,sha256=ODJe8YfNRP-hHsomFWk4_QpcuiSsNfjzGf256EDS0Pc,15802
590
603
  sglang/srt/models/chatglm.py,sha256=cajLN9caBl09e0TwOFkiTTKDqwlbmHo_yS-NCjdeQW8,13957
@@ -593,8 +606,8 @@ sglang/srt/models/commandr.py,sha256=5Y_b3K0QY7D37nFGkyiGgY38RleRui_GJUYcHSuHUZo
593
606
  sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,15598
594
607
  sglang/srt/models/deepseek.py,sha256=ZnN02HdgXCB23Vno5V9UMUoOxH5HC82vNTwsVulUJ-o,17206
595
608
  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
609
+ sglang/srt/models/deepseek_nextn.py,sha256=LaNBVtTSSO6_v5VRq-aZQ7K7lVFSIlUm8NcBB9p0fmw,5751
610
+ sglang/srt/models/deepseek_v2.py,sha256=8sYA5m2a590tF58IvRzOnPQUshKFAhQxQCbRmqEzoyY,89180
598
611
  sglang/srt/models/deepseek_vl2.py,sha256=j8BdxZsMjm6lPdbDipEIKhVIVywCP1Vl1Kl46BZ5_0Y,13147
599
612
  sglang/srt/models/exaone.py,sha256=TpO-rtCpEZ8Ua7hGFnS8l2oAYhY0Pij50grc9WQ2mvc,13576
600
613
  sglang/srt/models/gemma.py,sha256=4cdrPISg1VKnsuI-QPTpYvet4BrX8BMKvCIN82iLskw,12641
@@ -602,6 +615,7 @@ sglang/srt/models/gemma2.py,sha256=kqtwdo93GWKm2iBN29RoIRH2ggRm-K_80LM5btgfBLo,1
602
615
  sglang/srt/models/gemma2_reward.py,sha256=V8U3_ADUHWPdOwvEe1jhGW-oJmBgL8t1TY3-67Ksv2A,2618
603
616
  sglang/srt/models/gemma3_causal.py,sha256=Vm605KeF7CBXbtxzOWF-v0TLbl0G12CLF-bEgTV9T0E,25197
604
617
  sglang/srt/models/gemma3_mm.py,sha256=b9YmkipsfVb5IXVeIVwW_PviXiCkRULhEsqNOvPoDxU,17221
618
+ sglang/srt/models/glm4.py,sha256=2VQzUqFkQTy_2nfkxP9SF6_9kKLTZUExGRjge7r99Es,11265
605
619
  sglang/srt/models/gpt2.py,sha256=kclhxEs8oJk1KCyhmAqo7rZqecVGGHYkc-a1WZi3aIk,9841
606
620
  sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtYhl4,10307
607
621
  sglang/srt/models/granite.py,sha256=5WOJyNYAlt5RNHSexNfPNihhSxIMd7wPzju1cTixKig,20852
@@ -622,7 +636,7 @@ sglang/srt/models/llama_reward.py,sha256=LF2nqMV5XOrljGjAwJg43mBv3z6Q040I2EYlgZe
622
636
  sglang/srt/models/llava.py,sha256=m0-qcEWEUyN7kIiCYJAZTi58MwL4ioqCAD99PSTpsuY,37496
623
637
  sglang/srt/models/llavavid.py,sha256=q0lHlRnoYHKJZsWnkIQdd6dYAQ26t7XsmrqA0zDGmZc,12829
624
638
  sglang/srt/models/mimo.py,sha256=Mp-iFp4YHuiuq-H8enUF5K5QbMnVcvEa6mURH6vM3yM,6140
625
- sglang/srt/models/mimo_mtp.py,sha256=y0wTNObGz0lyVt0zu7apdVlTRLEQmc_xsd8SvgSo_vo,7806
639
+ sglang/srt/models/mimo_mtp.py,sha256=jSmqJAXu7G3OO7jW1oa2suI4H_Yl7u5ZT7w4lHFbHhg,7292
626
640
  sglang/srt/models/minicpm.py,sha256=m5HFsSJj0Po09LY9R6qj6K4gceqWDMOePz3NDGgMGT4,14691
627
641
  sglang/srt/models/minicpm3.py,sha256=6-gfHSfXwyB5zw7AIj-c5TzjpEKR6CM0H67MrD8LVUE,19347
628
642
  sglang/srt/models/minicpmo.py,sha256=IBMFTmak4qokNl8zBXHLYfS5slPf88ECwox6CKHbiAU,76081
@@ -658,9 +672,6 @@ sglang/srt/models/vila.py,sha256=DRGJXP6iSv0XjgHYIJpMzBDNYZb3__Gg_wFkmZ5D66Y,986
658
672
  sglang/srt/models/xverse.py,sha256=DsNVI9JpzN4jj0Ry6aTrj7r-xq5YLOoDX2kH4YLJA-I,14035
659
673
  sglang/srt/models/xverse_moe.py,sha256=7KCM2-j12towDMNvXkuuYiBOmNauH6NG4Ip40x0khqA,16782
660
674
  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
675
  sglang/srt/sampling/custom_logit_processor.py,sha256=AwODYVJdRkcQ8PGtJrhzKsqAgn8XZLQbAmR9fGiQzmc,1608
665
676
  sglang/srt/sampling/sampling_batch_info.py,sha256=JOGU-Fa8wPtytEJ30BHg1PldeNytwy8u31azLH436kw,13450
666
677
  sglang/srt/sampling/sampling_params.py,sha256=L2cU3MhAo9OvfdEtTdJidp9nTOkD6tiwlegxQ5vgIfU,6127
@@ -670,10 +681,10 @@ sglang/srt/sampling/penaltylib/min_new_tokens.py,sha256=rdU_D7RoIcrQPhysNQEzmr4T
670
681
  sglang/srt/sampling/penaltylib/orchestrator.py,sha256=XM-Lm6u7gYPtMZrTIc0FR4QxNZxBH5s_Cj82umyCzYk,5721
671
682
  sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=NRh10AJrrQlGJ6S-enGdRefrTrWpyqrSm-aNnyqQNQQ,2119
672
683
  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
684
+ sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=fSNaHYJOQUgy2_AHRlZPYSQRIMgSiL_Pz5h53eothx4,14239
685
+ sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py,sha256=sdIHbcWwgsXDoo64-T5veDuYB5eHC9QBwu2Ns29SCSI,14930
686
+ sglang/srt/speculative/eagle_utils.py,sha256=cBQYqryCr4is5E_16xIcUue0Q2_Mn4meY8NCyuu4RAg,45558
687
+ sglang/srt/speculative/eagle_worker.py,sha256=fjYJmzz8vYCf3xod0dYoGmwyFWh6iTjnGfbV6SdY32M,38105
677
688
  sglang/srt/speculative/spec_info.py,sha256=rhaKG0TzyF9XZYHEWp1jccwTBohSNsUDvxHFtAoOl18,709
678
689
  sglang/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
679
690
  sglang/test/few_shot_gsm8k.py,sha256=qnEejCyPHGEMBEiNc7T4-t9lvoo2DNN72RXw-ei_TnY,4300
@@ -690,7 +701,7 @@ sglang/test/simple_eval_mmlu.py,sha256=FkwamjGMjueTixymkedF-YiPloSLiy4ftILFUrKZ9
690
701
  sglang/test/test_activation.py,sha256=GeTIJHxlLQfW3kM-X1FGa8Sa3dSGKHEXl5wEy-hfGis,1489
691
702
  sglang/test/test_block_fp8.py,sha256=1qBD-FJr5UaR9e5qDz91vDrS_DT1utnk82qeQ2bm-1E,21630
692
703
  sglang/test/test_block_fp8_deep_gemm_blackwell.py,sha256=Hnhq4kkyINHb4ONedkp5Kf7Xx0MjOoOBYL8J7ETvOVA,8180
693
- sglang/test/test_block_fp8_ep.py,sha256=TI0TVeoX8zRDtOFclonCfgSH6vnAIOGiMKQKpf9aKNs,10845
704
+ sglang/test/test_block_fp8_ep.py,sha256=Ufq6p32sb1RSbFKaSnsVF8RrqqwwAACBUgOJTEH2Lbc,10856
694
705
  sglang/test/test_custom_ops.py,sha256=2bSo9P5_rJZYFq8Y8IKRimDfFyZZGJluhL7Ngny0Pf4,5571
695
706
  sglang/test/test_cutlass_moe.py,sha256=beXdvpwpHh0O8fpLW-RV3Bx8Wu0lRFG_SNkL3Op5ITA,9905
696
707
  sglang/test/test_deepep_utils.py,sha256=749ysTBGNzh6rYUCJhhZBtZpeD15eWTeNHYCytcvZtc,7448
@@ -698,13 +709,13 @@ sglang/test/test_dynamic_grad_mode.py,sha256=L76yUCuk_ymNpXD2CmO8r2GiGjIvD_gtTsu
698
709
  sglang/test/test_fp4_moe.py,sha256=U4LHzpGl6pCiCZiENRPxVP6hxqkqRZ_0WXbB29vYeAo,8187
699
710
  sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg,3764
700
711
  sglang/test/test_programs.py,sha256=DJIbTChloPwaDiEWHQoSTuKv3BGx0-K8Y5e1mDDTACM,18910
701
- sglang/test/test_utils.py,sha256=xdi1-Fvs0ykmaYhYzC0Jx-n7nSFKSO9a6HAvWbXqVQo,36176
712
+ sglang/test/test_utils.py,sha256=UfVq6snSqzomW0sDTql54Vbefl6kVovbFx1R79y_mqw,36243
702
713
  sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
703
714
  sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
704
715
  sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
705
716
  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,,
717
+ sglang-0.4.8.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
718
+ sglang-0.4.8.dist-info/METADATA,sha256=8Spz4kOLcrEbQuzVTiI4EE0A4Ldrdcholyuz3mNVUC4,26603
719
+ sglang-0.4.8.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
720
+ sglang-0.4.8.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
721
+ sglang-0.4.8.dist-info/RECORD,,