sglang 0.4.2.post3__py3-none-any.whl → 0.4.2.post4__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 (41) hide show
  1. sglang/check_env.py +1 -0
  2. sglang/srt/constrained/outlines_backend.py +4 -1
  3. sglang/srt/layers/attention/flashinfer_backend.py +34 -41
  4. sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +18 -3
  5. sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  6. sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  7. sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  8. sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  9. sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  10. sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  11. sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  12. sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  13. sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  14. sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  15. sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  16. sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  17. sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  18. sglang/srt/lora/backend/__init__.py +25 -5
  19. sglang/srt/lora/backend/base_backend.py +31 -9
  20. sglang/srt/lora/backend/flashinfer_backend.py +41 -4
  21. sglang/srt/lora/backend/triton_backend.py +34 -4
  22. sglang/srt/lora/layers.py +293 -0
  23. sglang/srt/lora/lora.py +101 -326
  24. sglang/srt/lora/lora_manager.py +101 -269
  25. sglang/srt/lora/mem_pool.py +174 -0
  26. sglang/srt/lora/triton_ops/__init__.py +7 -1
  27. sglang/srt/lora/triton_ops/gate_up_lora_b.py +170 -0
  28. sglang/srt/lora/triton_ops/qkv_lora_b.py +5 -5
  29. sglang/srt/lora/triton_ops/sgemm_lora_a.py +2 -2
  30. sglang/srt/lora/triton_ops/sgemm_lora_b.py +2 -2
  31. sglang/srt/lora/utils.py +141 -0
  32. sglang/srt/model_executor/cuda_graph_runner.py +4 -0
  33. sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +1 -0
  34. sglang/srt/speculative/eagle_utils.py +64 -21
  35. sglang/srt/speculative/eagle_worker.py +1 -0
  36. sglang/version.py +1 -1
  37. {sglang-0.4.2.post3.dist-info → sglang-0.4.2.post4.dist-info}/METADATA +4 -4
  38. {sglang-0.4.2.post3.dist-info → sglang-0.4.2.post4.dist-info}/RECORD +41 -24
  39. {sglang-0.4.2.post3.dist-info → sglang-0.4.2.post4.dist-info}/LICENSE +0 -0
  40. {sglang-0.4.2.post3.dist-info → sglang-0.4.2.post4.dist-info}/WHEEL +0 -0
  41. {sglang-0.4.2.post3.dist-info → sglang-0.4.2.post4.dist-info}/top_level.txt +0 -0
@@ -5,12 +5,12 @@ sglang/bench_offline_throughput.py,sha256=vIoF87HIpezB1x-xWzUl6SdXi88Fza8g4hDU7G
5
5
  sglang/bench_one_batch.py,sha256=d-LuRHEyDZjh180OCN5fqTjr8Zusk3zc0vhoJ33x0B0,17905
6
6
  sglang/bench_one_batch_server.py,sha256=iu73SsvYwnuRktYZDz1P6psMiRx8MbEbF5sbsYJdzYg,5962
7
7
  sglang/bench_serving.py,sha256=jYU3rYIDkzpYhjSpJw_IkEs_UNQfouNW4phs3z5TObc,54303
8
- sglang/check_env.py,sha256=4OqpZaEJOfBM6-vtPILto5kqDmgiZM1Koc7lK78A7CI,8427
8
+ sglang/check_env.py,sha256=lDVA3ybt1wOE33HIMpkkU7zGRgLWez1_ifRRJ8qxbtw,8445
9
9
  sglang/global_config.py,sha256=fnT0U9vlHdGaQFKN9tYTnUF4-eVW4HYQURd5zvPtrg0,1286
10
10
  sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
11
11
  sglang/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
12
12
  sglang/utils.py,sha256=7HpOrPBhMivWH719m7Dy1rjrAXOAsnqelpwNBBbvjqs,13319
13
- sglang/version.py,sha256=08dwZ-8Pb-Ir0QXBY3R8hBlzHyVuy4icqVMBMJri3oM,28
13
+ sglang/version.py,sha256=4aRfll4rDDWBVERpTNftyHYiEf_Z50L3jvLng1lO-j4,28
14
14
  sglang/lang/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
15
15
  sglang/lang/chat_template.py,sha256=v4SyYViPHX3i3XT46F7vlARn4UaSiP3PBpTGtzO6uRY,17006
16
16
  sglang/lang/choices.py,sha256=-W1DVw9N9ZliVpvmWrzIXG4cswAah8eMQrHWzkS3D8o,6234
@@ -46,7 +46,7 @@ sglang/srt/configs/load_config.py,sha256=la2ezNRcUZs7qiTYta2KEXqZ0U4TcmWW3U0sjoH
46
46
  sglang/srt/configs/model_config.py,sha256=sQIOfslBRzhOjucZdd8zE8nO9PEOc7zc6cZMbguQgoY,16876
47
47
  sglang/srt/configs/qwen2vl.py,sha256=ZjLy9v2eZY4wptUfY3CWgYKg2B5DDrkfCSyTy_Zf_bg,4351
48
48
  sglang/srt/constrained/base_grammar_backend.py,sha256=JFQFiAZLSqV6vck-ewIEzEEyncWLbRz_gkvkqpC282k,3185
49
- sglang/srt/constrained/outlines_backend.py,sha256=kNeJzr2NRqBy8AMiUlZUly0FmpR0raMB59QRhlfY_Tc,6970
49
+ sglang/srt/constrained/outlines_backend.py,sha256=yPYgz44n-rSCStGGkS1lGazFiQzN7gqwSvpJ2YG0co4,7081
50
50
  sglang/srt/constrained/outlines_jump_forward.py,sha256=iZWXeR3gNYoMubLGyFmLPO4V2YsN5DiGjD71Xk9iFaE,6418
51
51
  sglang/srt/constrained/xgrammar_backend.py,sha256=l-37tdrPsp7xnxZpY8_0W1DnZSiBAH9e-BcwiAO8b0g,5048
52
52
  sglang/srt/distributed/__init__.py,sha256=jFOcyt-wFAPMBUAf9zkZalNQlt-4rqmT6pCKBz1E4qo,149
@@ -77,7 +77,7 @@ sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSa
77
77
  sglang/srt/layers/vocab_parallel_embedding.py,sha256=txcjkuSDa6gZwESKj8X-HSLhAnMmDXL0FmFWY9SKqik,22155
78
78
  sglang/srt/layers/attention/__init__.py,sha256=KlQ0fl-o9v_NxBDhNZ4dPW2uQ2HeJjLm-0MTMWgaa28,2980
79
79
  sglang/srt/layers/attention/double_sparsity_backend.py,sha256=4mVyFPfZxPTwkQHGNCfI_4hQ8CbsWXJfxz-IQW77gAc,9143
80
- sglang/srt/layers/attention/flashinfer_backend.py,sha256=9BJEAQ5IcSMGvPfa6_D3cP9Gbo2XQ5GHBnF7cw2Rsng,42933
80
+ sglang/srt/layers/attention/flashinfer_backend.py,sha256=BlirMAu8dDicGdpZiW51hx9FpiU3EY-MNhZ1vdCJxGo,42398
81
81
  sglang/srt/layers/attention/torch_native_backend.py,sha256=KrcAqTLVZLtwgOmB0xhwUUsX32M-5LYZpNxaRNT4VuA,9252
82
82
  sglang/srt/layers/attention/triton_backend.py,sha256=mtLs768rhtCF_BVAV_rmYac0U4R1_HHc-9ic4JratsY,10100
83
83
  sglang/srt/layers/attention/vision.py,sha256=zLjKmzUlkgq1RFcP3b4EPArOAKovoaDLgYfM5SyB2wM,13181
@@ -91,7 +91,7 @@ sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NM
91
91
  sglang/srt/layers/moe/ep_moe/kernels.py,sha256=wb_S2qLxoWWgQu9coXy0XLNGvHzdZSdwXr0PGy4QySg,10940
92
92
  sglang/srt/layers/moe/ep_moe/layer.py,sha256=aS8t1XUvlTnO9IQaxGjW5bOXP4FrJDXzymEIvlIDMro,22603
93
93
  sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=h9yMFAL_bagUf-qBED8gSWdCOb7d8IdA-pE-L_nIg8E,842
94
- sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=tWV490Ao5vIasPDBBY9ktuAZdWlONnnv3uPCifcTfpI,37241
94
+ sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=NckXo7tQ0rp3trdAq5TB5mP4MmCdjm6RIN157d9Efvk,37720
95
95
  sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=-49WRpq9OtRZocQjW-YNcB_ruK09nIJqGHKNa8CJsws,22691
96
96
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=iNGsE2ZeVnQEnN4A8UJ9Jv0d3hbRF2MJ9oBgjup5Szk,2737
97
97
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=JJN0hryyLr5Zv3dSS7C8cPFhAwTT6XxUVnBGMZvV6JA,2752
@@ -202,18 +202,24 @@ sglang/srt/layers/quantization/int8_kernel.py,sha256=t_BLVf8XjOyn7S3Lu3B4hXvw8Dv
202
202
  sglang/srt/layers/quantization/modelopt_quant.py,sha256=_VdVz77dTP-IczPeFrdH6Ttro2D26BZvMlZkCKWj_5o,6200
203
203
  sglang/srt/layers/quantization/w8a8_int8.py,sha256=PBapNo27rkUdX95E0ihUZeHTXP2Id3CwlNlMU8GIuZc,3344
204
204
  "sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=tkLjwLC_aVXhzuvo-2QHkojXZauPJsf3jNHFn1S7uRA,3244
205
+ "sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=CPo1WRF0HgsQMPBkvpoImElQMrfwpJLhEvL86e6fkPU,3247
205
206
  "sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=xqJNygFgPCe4jDpkfVOWCXpuNMUjmssdD8oGhp2iXv8,3726
206
207
  "sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=xqJNygFgPCe4jDpkfVOWCXpuNMUjmssdD8oGhp2iXv8,3726
207
208
  "sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Qoj9rLLRDbKM4IKBCXvN8RcxzSmNPd0TQUiM7CXDqHI,3241
209
+ "sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=UZljnxxCSjwnZlX3OgKWZJGXCf5BWF_agEpNX8I4Zxc,3248
208
210
  "sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=4D3Ku4y7BCVEJzueKvQC_KvOR026w3ONWsxfsA_YrEc,3249
209
211
  "sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=7v4tp0RaT4vxF4urSBrkK5FR_5ikeFQ1htF3DwDl1lk,3249
212
+ "sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=3matoCzEe4aexwoe7YTmkjyE4NA8khWXjL5EySuNwzA,3254
210
213
  "sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=0J2MFgaLkv-mfVE5x363lgVKYU6miLG_xRO3tJUga_M,3249
211
214
  "sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=4ubbhwSFX_XbefRLEkLoWxJkcetFWPzsszPu0X3_Wrw,3242
215
+ "sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=9f8Ib4gLEFSfdNpO8IL8uiONImvqnlPbJrZ0HM3OB-o,3247
212
216
  "sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FhyniGTx5QeCuVrBSVTQys6q05Pr5lPEcPykpAX7Iyo,3247
213
217
  "sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=0v17v78pETXv6S2ZoibekxOVhiTmCm807DYG4DONUck,3259
214
218
  "sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=I44PvJj758-sw_fCOVROLTpG0NQ5_5PCYyQcpZC1YSY,3259
219
+ "sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=tG5_iVeRBHTgHX-liOf79nWRjj_lUZ-NQWTbBrBgORQ,3246
215
220
  "sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=-CVHqClROli9FWe_FnlnuAG2LiFivDFK_nghH6t-BWc,3261
216
221
  "sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=GsLoYkaZ2p4Qu0Coj-X90s7JWyfZBOloIHPlyNKSIes,3246
222
+ "sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=BpKweY4HfBx5xvqSvwNahy9x7R5FH-YK8j6rLFQdKwo,3248
217
223
  "sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=I6a5wQ0R3cBp8ix_PDamWZN2aJmV_1p0tg430L3Updg,3727
218
224
  "sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=I6a5wQ0R3cBp8ix_PDamWZN2aJmV_1p0tg430L3Updg,3727
219
225
  "sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=8zuJhFdd6aXREpiqPFhIKEFWA5lgLVGrG0-a9UXcBqk,3262
@@ -223,6 +229,7 @@ sglang/srt/layers/quantization/w8a8_int8.py,sha256=PBapNo27rkUdX95E0ihUZeHTXP2Id
223
229
  "sglang/srt/layers/quantization/configs/N=3072,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=M3nwpZd2-0w263ywZt9gaw53z7MN673T5tl4tc43Ntk,3249
224
230
  "sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=vLoV3JMtvHOKpR5D1BeCQPMuYlWUAlrXu54gByNkwKY,3266
225
231
  "sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Mtw7a9BSspj2TzC-aPxE82o1LEvwzgbUuIofwRxUNA0,3263
232
+ "sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=B0lo3SuoQXhBEnojH2TwpVeurvlKD8yI8kQrJ5ORhWU,3249
226
233
  "sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=NHdx3tZnfLF7NplswMzcTRbQEQFLtChg4rd7GU9lMbM,3262
227
234
  "sglang/srt/layers/quantization/configs/N=36864,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=laYeH4w0iZOj2Yg3vDgtKoroNQnwBEX4GUGLrO9095I,3260
228
235
  "sglang/srt/layers/quantization/configs/N=36864,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=TWcPDZ2miQMD6OWDC1FteRs80ND9RC-oJL3PLVmJbtI,3257
@@ -244,17 +251,23 @@ sglang/srt/layers/quantization/w8a8_int8.py,sha256=PBapNo27rkUdX95E0ihUZeHTXP2Id
244
251
  "sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=xsFMrq4aybClfJyhm78c1Hf1jcyFSGnfygdHYp7OhSQ,3727
245
252
  "sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=qG6v3n3qF6LE2DdGT-mDIXecZ1a7vg7p3QqXYCMX85k,3254
246
253
  "sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=EgFTGyW_YuDwyEDUCoGglyI1ETdj9J7AR0UfJ86jMoI,3249
254
+ "sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=BpKweY4HfBx5xvqSvwNahy9x7R5FH-YK8j6rLFQdKwo,3248
247
255
  "sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=NiorJgOotxkQcP49ID3z5al1UA4QQDrT8MvbCwAWL5Y,3248
248
256
  "sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=bPQWtvaJrzOOIgI-R-MIxs_f4yC_FobkDydu3OkOFtg,3252
257
+ "sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=h32lCTFilLlyKbMeuJvNWG1v0yJJzNj93kwSvlrHfaY,3249
249
258
  "sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=ZRgiuHZ2SFC6u-WV5DGwau4k1RiPLI67eENO0e-5Ylg,3253
250
259
  "sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=-hP_P8NM0K04mGzTmpGBNibQ5xxh5gPz5WtoMXhoz1E,3253
260
+ "sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=0jX-z2lTgVw7ABLmWsIsQdqW4EjmbXKRDHye_XPLCAE,3245
251
261
  "sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FB5Le4obvPoCgFSnC_3-Uh59n-Mt4Rol8saXVcK3RPw,3252
252
262
  "sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=kLviGvVngpgOuelfKtvv9Is7MWQ89rGxlomMRP6t0Ic,3250
263
+ "sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=_exM3wJ3FMmGHweBcH-8IxwZBzaOmPaF3ScMM6KDpiY,3253
253
264
  "sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=vlys0Zi_CaaU41OHGbWSBtbVglFi98bgqEySBMc9Sdg,3258
254
265
  "sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=YWyByOlKSqp5lbcUa8eu6N2dHRKJqJDbCDSjdDQJngg,3249
266
+ "sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=ENRWYdUwI0ooHb6IwcHliupRWOPnw-7-WtxZB-qQGJI,3245
255
267
  "sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=j5PTW0IC4Z2yQIygcdICaOsvb639u6Mv-ZpJYkrBQ2k,3254
256
268
  "sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Ggy4hejkcWjiw5Bi-wGzSP5JLVuvOjip_rbjXFBJZbs,3257
257
269
  "sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Xy4mgZx5iiEvuv2ydO4dFNIT8s0jgBhNHE1vu93fGJM,3250
270
+ "sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=TdWuE2RIsIyr4Im24MuWK3XyiNtbhO_hAiAXDz5gNUk,3246
258
271
  "sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=x476nFeltB_2iO9_6y-z2P_unAbh7ghLPFi5z2LOTOo,3253
259
272
  "sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=sVbH4YRLTxBqvTh_6xbtXkj3orOrKytlwM-_4gtD6IY,3725
260
273
  "sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=sVbH4YRLTxBqvTh_6xbtXkj3orOrKytlwM-_4gtD6IY,3725
@@ -270,17 +283,21 @@ sglang/srt/layers/quantization/w8a8_int8.py,sha256=PBapNo27rkUdX95E0ihUZeHTXP2Id
270
283
  "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=aoq4792zPo87QO7VrEf9fb_vj0zPiHIu7Ho9aMXwcLw,3731
271
284
  "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=PD4AJYCkHfy2ivv9baMouFXzBTy0eKMumbAfxfm91HI,3256
272
285
  "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FFBjSWlpKXMxfAUUYUqXbOK_Hd7qBeBsfbcaa9uB4qY,3249
273
- sglang/srt/lora/lora.py,sha256=71sVFPB23THA4O06v2_G91vaY3MhAchsti-LErycJ6g,14869
286
+ sglang/srt/lora/layers.py,sha256=r34oprzwyE3SWPvaNkBvXWPtfa-0IY987_bjj36ySfw,9996
287
+ sglang/srt/lora/lora.py,sha256=_WrZxS6-sarwUPvumcReyKGrH6fSCd8-UsoX56aQJ4s,7293
274
288
  sglang/srt/lora/lora_config.py,sha256=a2fTQESlCbG1xLiBYy4ptZ6c0Burcqyg1_6V1XSok-Y,1506
275
- sglang/srt/lora/lora_manager.py,sha256=fa-Sy4b3k_YzJJY5XQgxCX5AycnraF-f600jDBaXKzo,13991
276
- sglang/srt/lora/backend/__init__.py,sha256=ul5yTkHrwDTv8RZXu_Miw5_XpBlxqKeGmLfUT6Ph7EE,211
277
- sglang/srt/lora/backend/base_backend.py,sha256=MZfUhIwujwDyN2hFoVMB3N_gJokel6aU4Scc8fv5_L0,3575
278
- sglang/srt/lora/backend/flashinfer_backend.py,sha256=qXqtt4w_-Gqkk-j27CHHI1Jeu0gb8jr0ngI54eCNdqI,2845
279
- sglang/srt/lora/backend/triton_backend.py,sha256=MzVZYtPXwqS8DfeFMIpI0xNI90oVHcasaT3ppea9Vns,1722
280
- sglang/srt/lora/triton_ops/__init__.py,sha256=F4NrXZ4bSD39hkPVkby6OR-0Ap7dOOUsB0BjqjGMjy4,195
281
- sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=gFO53wGm93GpDmFaJgqyD1tgmrDllZ2bCBhGczZg0_4,5664
282
- sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=GvPFXGXC2qLh4djlIIyAvobo9VLJcSEtyjcaveQh0Cw,3981
283
- sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=pZxVJOMaf0OwMUJAw37KA4ZcoQb8gLpFGbYPirX8Oqk,4298
289
+ sglang/srt/lora/lora_manager.py,sha256=-7ZWAL-E2mW3acXd9M0Z_slnTV1GfzGBD4RRv3cjafs,7851
290
+ sglang/srt/lora/mem_pool.py,sha256=eV_GXETxNODPVIAnTEeUUUVn0IVgguBR_mYFzIK-VHA,6835
291
+ sglang/srt/lora/utils.py,sha256=6i7Q1Y-1LLbRkeCMv_lKIzkTN0veUTLbc8wlHn7R-bA,4571
292
+ sglang/srt/lora/backend/__init__.py,sha256=98L_KRRnE3gcGcx7Lb6yjAEUUE_Yay3QszcQXdzYsDw,708
293
+ sglang/srt/lora/backend/base_backend.py,sha256=dldwA7vTWrB1ln1MwLYKNtMkBoAgD7OLSlWe9tL2lzk,4602
294
+ sglang/srt/lora/backend/flashinfer_backend.py,sha256=fXfkl7Cpw8ap2bCrgWdn_gEUzMXX1pNjNuiPw3kA76U,3984
295
+ sglang/srt/lora/backend/triton_backend.py,sha256=ZT5M30vj8x77Kltukpga4wk1sd8fT4n_FdsOMQBTMI0,2610
296
+ sglang/srt/lora/triton_ops/__init__.py,sha256=JGOYPIn1XbGcyJTbt8A0qoc02PYONSGNNjGkC8yJpAM,283
297
+ sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=qve4oNZHYUFk9ckmT2BVuDNMEvrN7Quu6RsS8Iz3uRQ,5066
298
+ sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=BmIcTZMnlSnie9rnMl4KvLpc4Njsk7_IppbUqitf9Xw,5738
299
+ sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=kv-AvJ_Bi3yWjGvFnSwXvP66iJvY9n9pEnJzJ9-DWzo,3982
300
+ sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=Ai5vPriT4OgACwK7xrpGgf5L1oaN9x0jwNKMChu3uI0,4299
284
301
  sglang/srt/managers/cache_controller.py,sha256=DXnIunJgtTws1WF2vZOYVQe56vacV7Mn4wL9zoG8Xz8,10909
285
302
  sglang/srt/managers/configure_logging.py,sha256=aY9xExurz7t_IdItd-9GuVuM7kEGB8_bRryhZxKdu9o,1542
286
303
  sglang/srt/managers/data_parallel_controller.py,sha256=b64aC6iLr5RolJyNQnT-yTQ_TSI9DDLtuABf_TPTUrM,9421
@@ -302,7 +319,7 @@ sglang/srt/mem_cache/memory_pool.py,sha256=9ud97u1cXnN6O0qlR8tv8woN_20gqisTV6aBg
302
319
  sglang/srt/mem_cache/radix_cache.py,sha256=hVILXvc5PauHuLTeyZbm3NCf3AOimaAuXjll53MSLeU,11754
303
320
  sglang/srt/metrics/collector.py,sha256=_yl0_paSARxS1ypZgd-pLJ29tMizolHuwROX21dOXTk,7326
304
321
  sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
305
- sglang/srt/model_executor/cuda_graph_runner.py,sha256=Tv9ktSMaHNcAUUU2ERmxouFc6c94qbNMGt7_usQCD3w,18202
322
+ sglang/srt/model_executor/cuda_graph_runner.py,sha256=7MLskhry5KFP-lhqUfAwZ8P4HckFCEQonYqIDjfAWSw,18455
306
323
  sglang/srt/model_executor/forward_batch_info.py,sha256=t1RlBgoeS-_Ikl28Xjvt-aouh1nNUc3eLM4iGY4_QqY,14988
307
324
  sglang/srt/model_executor/model_runner.py,sha256=5ET8CXwSrzIKb_rDW-6S0p7tv-iCPoEwZk3dESwPdF8,33090
308
325
  sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
@@ -366,9 +383,9 @@ sglang/srt/sampling/penaltylib/penalizers/min_new_tokens.py,sha256=_Nxv0XgUPirZj
366
383
  sglang/srt/sampling/penaltylib/penalizers/presence_penalty.py,sha256=5tOgCg7OvE9kSN9VMCpH1hwqo1YMxt9iS5PVpct9HpU,2468
367
384
  sglang/srt/sampling/penaltylib/penalizers/repetition_penalty.py,sha256=l1DyU8kC8n_F4Z6Jd8mZKfF23buuLZ5dWuVfyqDWkUI,2968
368
385
  sglang/srt/speculative/build_eagle_tree.py,sha256=zWthboIgzPzSOXcGxDpDv0rBOQP55HYGrBKGqm2gWF0,20732
369
- sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=5ZCy6ndPA2p95xDgo2kXWD3zCtVaq4q5X0HBpAbB3Xs,7929
370
- sglang/srt/speculative/eagle_utils.py,sha256=BV89f2CTp9H0pSvJfK13WYvTL7LW3BtcplQfLngKihg,24451
371
- sglang/srt/speculative/eagle_worker.py,sha256=4oROLwUBJIwEHNHNEfvsy74DqLQLVc4KfjdR-MrB1OM,12038
386
+ sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=FY4hcwd0Blx7AXbeX6quaXPNgWA8WGIqVcQiEgHyERk,8002
387
+ sglang/srt/speculative/eagle_utils.py,sha256=ypjVmVTVzCGclOVHRMJxdLUSPkf1-7bNXQS0oP6dn5U,25644
388
+ sglang/srt/speculative/eagle_worker.py,sha256=yDCU4liKU_jgtwQ-8L9_GuUtFRo3l1Df7mCeKX8yHCo,12085
372
389
  sglang/srt/speculative/spec_info.py,sha256=D7A27UU1iOwIBEjXTgAxZ7jdftbTiVlMCvK8GmYr2zg,488
373
390
  sglang/test/few_shot_gsm8k.py,sha256=7yDbEQe49gZeJhz2wFFX-gf_59ThDKsCS1xwfogNc7k,4034
374
391
  sglang/test/few_shot_gsm8k_engine.py,sha256=QQbrwOX6-cJDD3RZC_e7zPnt6aSo8JdF8X_lRHSjdDM,3886
@@ -386,8 +403,8 @@ sglang/test/test_layernorm.py,sha256=IacByD5d-stXjzBz8Ypamc7povlcedpKPbb_4JLgo3c
386
403
  sglang/test/test_programs.py,sha256=aUV9Ex_B714ph7ytv6W3J7sdGDKC6lGIhUy95Yg6AHQ,18878
387
404
  sglang/test/test_utils.py,sha256=BU6lAX3bu3TNQZqVC9UPnyq3I7iV5kigHQKJx7UNlOQ,26192
388
405
  sglang/test/srt/sampling/penaltylib/utils.py,sha256=CjxHgywh0hx_87iynzQt_ztHu6zBVuE-YrZ-XPmW6U4,12906
389
- sglang-0.4.2.post3.dist-info/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
390
- sglang-0.4.2.post3.dist-info/METADATA,sha256=eVi6WuPieNGNX7TzNcBd8JolIIPQphw6609pqdALCUQ,23763
391
- sglang-0.4.2.post3.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
392
- sglang-0.4.2.post3.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
393
- sglang-0.4.2.post3.dist-info/RECORD,,
406
+ sglang-0.4.2.post4.dist-info/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
407
+ sglang-0.4.2.post4.dist-info/METADATA,sha256=VylN7wgaeE_8T-ddwxTmujJo04JCRpaP_uyhGfo-InQ,23773
408
+ sglang-0.4.2.post4.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
409
+ sglang-0.4.2.post4.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
410
+ sglang-0.4.2.post4.dist-info/RECORD,,