sglang 0.4.9.post5__py3-none-any.whl → 0.4.9.post6__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 (32) hide show
  1. sglang/srt/configs/model_config.py +3 -0
  2. sglang/srt/entrypoints/http_server.py +13 -1
  3. sglang/srt/entrypoints/openai/protocol.py +3 -1
  4. sglang/srt/entrypoints/openai/serving_base.py +5 -2
  5. sglang/srt/layers/moe/ep_moe/layer.py +152 -37
  6. sglang/srt/layers/moe/ep_moe/token_dispatcher.py +83 -118
  7. sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
  8. sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
  9. sglang/srt/layers/moe/token_dispatcher/__init__.py +0 -0
  10. sglang/srt/layers/moe/token_dispatcher/base_dispatcher.py +48 -0
  11. sglang/srt/layers/moe/token_dispatcher/standard.py +19 -0
  12. sglang/srt/layers/moe/topk.py +6 -2
  13. sglang/srt/layers/quantization/modelopt_quant.py +2 -0
  14. sglang/srt/managers/data_parallel_controller.py +4 -0
  15. sglang/srt/managers/io_struct.py +12 -0
  16. sglang/srt/managers/scheduler.py +29 -0
  17. sglang/srt/managers/scheduler_input_blocker.py +106 -0
  18. sglang/srt/managers/tokenizer_manager.py +43 -9
  19. sglang/srt/managers/tp_worker.py +5 -0
  20. sglang/srt/model_executor/model_runner.py +15 -13
  21. sglang/srt/models/deepseek_v2.py +13 -56
  22. sglang/srt/models/qwen3_moe.py +12 -69
  23. sglang/srt/poll_based_barrier.py +31 -0
  24. sglang/srt/server_args.py +8 -0
  25. sglang/srt/two_batch_overlap.py +8 -3
  26. sglang/test/test_utils.py +53 -0
  27. sglang/version.py +1 -1
  28. {sglang-0.4.9.post5.dist-info → sglang-0.4.9.post6.dist-info}/METADATA +2 -1
  29. {sglang-0.4.9.post5.dist-info → sglang-0.4.9.post6.dist-info}/RECORD +32 -25
  30. {sglang-0.4.9.post5.dist-info → sglang-0.4.9.post6.dist-info}/WHEEL +0 -0
  31. {sglang-0.4.9.post5.dist-info → sglang-0.4.9.post6.dist-info}/licenses/LICENSE +0 -0
  32. {sglang-0.4.9.post5.dist-info → sglang-0.4.9.post6.dist-info}/top_level.txt +0 -0
@@ -10,7 +10,7 @@ sglang/global_config.py,sha256=xzLdk8W53fneFblNh8iIjGF9C3-7mnzR1-LleD9Btxg,1495
10
10
  sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
11
11
  sglang/profiler.py,sha256=tEHzHerXC-ymk4OrkoUcMbgcGHmb8VESthsNSP2Yx9w,4417
12
12
  sglang/utils.py,sha256=DSMD5Ch-gmkIO6Dh4JAmBMtRAu4JZ7uYO0NUMprGVO4,16770
13
- sglang/version.py,sha256=GBi71Z9Lv3isXS33OtWCtTjl5apAGantvWQRrye9bj0,28
13
+ sglang/version.py,sha256=sV0w9abphYZ3w1RIYSGHC1mPd8_FLDhrgT8ogSNOvck,28
14
14
  sglang/eval/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
15
15
  sglang/eval/loogle_eval.py,sha256=-CC2s2kh5qUoDrHRkQVkC_jNvBgNojXbf456ny5s78s,4557
16
16
  sglang/lang/chat_template.py,sha256=ZNcuDvDZE9781tQqfKpptck_zJVK0aggq7IxrNHltao,21934
@@ -38,10 +38,11 @@ sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,
38
38
  sglang/srt/operations.py,sha256=ddQ8KO63L73OciaR8MZ9h2h83gKVY4-WuWgeEGowPJA,5346
39
39
  sglang/srt/operations_strategy.py,sha256=Pwd2sKeRtKh9WJXgzlNr2tU9y6YMcI3MDLbatHqlMws,7145
40
40
  sglang/srt/patch_torch.py,sha256=OUPCGQSQz3MVZB1zZ_Eq8lXiw0uIKJ_HWjqQolI8FsM,3088
41
+ sglang/srt/poll_based_barrier.py,sha256=9vjn32ROlSueEk7EuZjWKUrrL7XVR9KJsSVliGTYB_4,937
41
42
  sglang/srt/reasoning_parser.py,sha256=6lkrkR8Q25EU3VAE0jjSq__361e3CkX0gJIieSfjsZk,9247
42
- sglang/srt/server_args.py,sha256=hfPgshfRHHq-_wI4jjMAjMGeXqmr3ZSBB_3ZVW7c8Bo,85937
43
+ sglang/srt/server_args.py,sha256=-FRh5ZtXHAN2PVthw_LcH0KOgzlzeohFU24ZkeZBVm0,86265
43
44
  sglang/srt/torch_memory_saver_adapter.py,sha256=K_eTx0UU84MHSTXI3iqYLdHV4IWtJMJ2FKdGFJR8v1E,2417
44
- sglang/srt/two_batch_overlap.py,sha256=Lx4bIXJb3hd9vzuJbmp3Crf4I8_NrT9fUbs-4LRzQXk,28711
45
+ sglang/srt/two_batch_overlap.py,sha256=AgaPeAXJQjt4RVhIVKR2YMEZRL8NqjN93NpgDE_fXfQ,28892
45
46
  sglang/srt/utils.py,sha256=6YkLoFlKyLpCNlk6zzaM5ys5_dMRTMOWNoLH09oEsOs,93321
46
47
  sglang/srt/warmup.py,sha256=zldxhMlXpclRAJXmfBjJNUJd1eDizVdysibBvQyTVuA,1782
47
48
  sglang/srt/configs/__init__.py,sha256=8EcVRP95epZ49DxBa6LgKWt7eO3Qe7Hrr3V1c6HkMnY,553
@@ -55,7 +56,7 @@ sglang/srt/configs/janus_pro.py,sha256=yI2c_aWBIKiTcK6RZxK6hq8CIxyYRH_NN8uaSNKsr
55
56
  sglang/srt/configs/kimi_vl.py,sha256=4W7VQI3pr888ZsFA2SqCQo4mI0seXTOrGQ-x3oTvWew,1358
56
57
  sglang/srt/configs/kimi_vl_moonvit.py,sha256=hx2Rt4JSFbvy2HUTeLjBpge87m8M6ITAhqsgdNf_Jd4,1163
57
58
  sglang/srt/configs/load_config.py,sha256=qs-AxuplouBx2tsv9KGBOLZPbwzuVA4vbktbGP_cRp8,3309
58
- sglang/srt/configs/model_config.py,sha256=X7xirRQx1M0OhM7L-cOPJqHfBFIOfiZiiov3E847nqY,28613
59
+ sglang/srt/configs/model_config.py,sha256=8lQ0tH3RORJAYSSosaS0jCPtuXkwcs6XUZsEQsZqgqM,28738
59
60
  sglang/srt/configs/update_config.py,sha256=D-myV9Ombwp6wTF0RX9qUO6m5rxFiUcyse32FQADG28,4787
60
61
  sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
61
62
  sglang/srt/connector/__init__.py,sha256=czLX5JOxuMhH-T9eSJzoc1qv1B4z9chyffDRL5I6wo4,1247
@@ -116,11 +117,11 @@ sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=IrSrnpZnii0E
116
117
  sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
117
118
  sglang/srt/entrypoints/EngineBase.py,sha256=yKN76witT2jz1zhmLHmPNLGMpK2UiOTaKQ2KPD8l99U,2594
118
119
  sglang/srt/entrypoints/engine.py,sha256=dGazuzKYxQSfXjea1aeSqaZubvLER0EVZzD7CvpHq9A,31595
119
- sglang/srt/entrypoints/http_server.py,sha256=-unKM0lCK7Scd3LLTxLl-3vhit6lMb-wWfjLywelRy0,37214
120
+ sglang/srt/entrypoints/http_server.py,sha256=aoXYyq5i4jfBppDrX5lECFFvsRlT90eW03gBXDTUpyQ,37647
120
121
  sglang/srt/entrypoints/http_server_engine.py,sha256=_--j4U04OeJLlnnv1f0XmCd_Ry0z1FlhkrbePX8rYV0,4938
121
122
  sglang/srt/entrypoints/openai/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
122
- sglang/srt/entrypoints/openai/protocol.py,sha256=UUc5oJ7E60fYoUqMb53H85i7HHhDfmYLIFZidhNrP0s,19147
123
- sglang/srt/entrypoints/openai/serving_base.py,sha256=5NJ2S_6B2NFSwn4nLp6eaeJ5iC3IcQzMEY9lW_gPcdA,5246
123
+ sglang/srt/entrypoints/openai/protocol.py,sha256=2XfURpoICsk4UZs4W9lm2A1yXNC-iwqSdZOuof0FXjo,19183
124
+ sglang/srt/entrypoints/openai/serving_base.py,sha256=epD6J7K72p6TP8o2dGYf-a76nXeF80vFhPRpWts6684,5445
124
125
  sglang/srt/entrypoints/openai/serving_chat.py,sha256=MbLFLLFhvCTpJ_j7t03seXHHoFjlXCZtzVjSg_L7iz8,39841
125
126
  sglang/srt/entrypoints/openai/serving_completions.py,sha256=emIutVmnJgOgGRuAqdB80-kVHUfQbE67n1VS-76o4QY,16690
126
127
  sglang/srt/entrypoints/openai/serving_embedding.py,sha256=hqPt2ELpT1yQ8sBXo801aLJ3sExoPXs_K5ZQtC2vpAs,6274
@@ -198,11 +199,11 @@ sglang/srt/layers/moe/cutlass_moe_params.py,sha256=9NRCmgP_Ug3gGqCcpi-x-QRbLjCNp
198
199
  sglang/srt/layers/moe/cutlass_w4a8_moe.py,sha256=bc8s5Oc36pCF0VozxZuqFOOU2fov-4wGPlyduye3Imk,7296
199
200
  sglang/srt/layers/moe/fused_moe_native.py,sha256=SkOFgU23aZfHmLhnkVhdUWGx0-2wbo6uu9ohKBp2yOI,3486
200
201
  sglang/srt/layers/moe/router.py,sha256=UrPieRvemN7Ew48gtG7DA2xhNDBRSnZxzugTEBI-0_E,12006
201
- sglang/srt/layers/moe/topk.py,sha256=dOyJcpjIQYlgk8i_JsJyJzxaZZgmAcSMr9OgceG0arQ,26851
202
+ sglang/srt/layers/moe/topk.py,sha256=ttPFvXzTY0eyh6UH8fJZCjyDp-R9EssSP8m229MxQ_4,26925
202
203
  sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
203
204
  sglang/srt/layers/moe/ep_moe/kernels.py,sha256=UFH-XacP3aBjAcn6_SeC641jGzjSrY0fz50IB9URix4,42706
204
- sglang/srt/layers/moe/ep_moe/layer.py,sha256=6fspJli5_89rItBZDua_gNMP8eAq9ZgFOv029aoIIwg,45701
205
- sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=aiBE3mjvWV3eBrFGH9J44tuJncQwOjRS_XeyBNCEtqM,24379
205
+ sglang/srt/layers/moe/ep_moe/layer.py,sha256=7VyB__QLkEJg-WLwxZ8yv9Qc3CFDw5HvmNlkKmQ1JFY,49501
206
+ sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=CJhFTtOydsLSXMYNfzAASPI5p_oP7c4krGwX4bBcT1s,22909
206
207
  sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=4NXZHbCw-G-uSnNUj4up0yh3xBDPnT-x0pdoIr0lku8,831
207
208
  sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=6NcY-6UHQUSVP4qzc8jis4h-W9BEV4lnGmrJMx2S6eA,57164
208
209
  sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=U3Qaffyj-aGFc6TvlB057eCbiv9oHbPDdwPVHsM7FJs,23278
@@ -368,7 +369,9 @@ sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=HOEJJyehKTUI
368
369
  "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
369
370
  "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
370
371
  "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
372
+ "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json",sha256=rqPWpRfa9s7USa9cUyHAVP50WYkCqXDJzfKZhqk6GHw,3267
371
373
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=320,device_name=NVIDIA_H20-3e.json",sha256=cWkEpNqnyn0QS8HcgWiwWI1xqh_U93_S1kyNeb69aOc,3238
374
+ "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8.json",sha256=jGWfZNP5T1en8KX0_94lqqTKRQXC_Kkq_lJrtY4gE8Y,3263
372
375
  "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
373
376
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=384,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=ENDYnC1ljPNpDVHdmqR7UbgDWT1IP_55ZIQ_Oae34mI,3264
374
377
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=384,N=128,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=5UAs8Z1myw8F-5FY2HHWd5WMT2ii9X-aYIYkf3FsA8s,3269
@@ -376,6 +379,9 @@ sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=HOEJJyehKTUI
376
379
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=385,N=128,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=rgF-KSAUXmI7QGrhMUywPVfeTxz_j8_IP4N-6bkYdak,3260
377
380
  "sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=385,N=128,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=7aaoi0MI2xC5ZJkflkTZAH4tawTPdmUpmcBOgOInKxs,3268
378
381
  "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
382
+ sglang/srt/layers/moe/token_dispatcher/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
383
+ sglang/srt/layers/moe/token_dispatcher/base_dispatcher.py,sha256=Ut-B7CmWcVPZCPodqBobIdfbclmgs03J3A-zAU3nyNQ,1095
384
+ sglang/srt/layers/moe/token_dispatcher/standard.py,sha256=jseBfZpst5iCDxXxQV_-53qb0VMqy8RHqXJMf86Bdgg,430
379
385
  sglang/srt/layers/quantization/__init__.py,sha256=lMjOmhMntbY55RR5eKKLB338tdzlxGuBx-pT4f9jmIM,8130
380
386
  sglang/srt/layers/quantization/awq.py,sha256=gLEOx6QoGQ0McBEURXPVidKGsdNUjqc9sJwGLQx86rE,26690
381
387
  sglang/srt/layers/quantization/awq_triton.py,sha256=4XwJbRNlFmlvjbMGtWF1Wvk26Kf96162vNEmuI5Kars,11737
@@ -389,7 +395,7 @@ sglang/srt/layers/quantization/int8_kernel.py,sha256=GfRC9FOn9exNvK4QHbUeBj3Hhv3
389
395
  sglang/srt/layers/quantization/int8_utils.py,sha256=YK9CS-lb_n91kNCTKK5o5apYF31V2giDg5G5VKrpcUA,2356
390
396
  sglang/srt/layers/quantization/kv_cache.py,sha256=_9pF5rwvB7ta6Gdc5YKVVGbNzYwqmhIx4TrX1-xnodQ,3261
391
397
  sglang/srt/layers/quantization/marlin_utils.py,sha256=ShsoxcVTYo1Es3XKp9mlQNw-TsRF_o4mHIO5MfAP5HA,25906
392
- sglang/srt/layers/quantization/modelopt_quant.py,sha256=pJDavvIiy6oY__GoknDgeq1o8jVF99CzYgWy7i6Cz9E,38288
398
+ sglang/srt/layers/quantization/modelopt_quant.py,sha256=oj6jJ66s66kMCufuJapUxGaAX8PjuJB2pdzZuUDZcrM,38357
393
399
  sglang/srt/layers/quantization/moe_wna16.py,sha256=NGcFjiMXwfOX10dkHAY3EgZHxox2k7I402Bvwj3jW1Q,19105
394
400
  sglang/srt/layers/quantization/petit.py,sha256=74Jn0eICq23v_b4rEctDaYTINtbqkdM6IXycT-_FbbI,8954
395
401
  sglang/srt/layers/quantization/petit_utils.py,sha256=-gy4zMhqNoIA1R0n7-5C0efV54jHonCUgPDUUFjGsyM,3245
@@ -581,19 +587,20 @@ sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=Ksova04wgeGsFqGOXWqJtMYaHgyUYc
581
587
  sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=VqCAFvUtq_l-0RGIkx3W_fzD55QcW2FAcqpvSHOBFp4,5115
582
588
  sglang/srt/managers/cache_controller.py,sha256=HoI6_os2h8rrvyEJ6Kk4ibUF3aKH-9Z8K5pksc6BECQ,24603
583
589
  sglang/srt/managers/configure_logging.py,sha256=8sNXZ2z9pBWOwn-X3wyz013Ob8Nbm1zDxRkxoZjH-l4,1633
584
- sglang/srt/managers/data_parallel_controller.py,sha256=PZ-wOcAFn3PQqMB6I2vjIsFmplf0nlNl5hqTDKTHTG8,12112
590
+ sglang/srt/managers/data_parallel_controller.py,sha256=gcg_TGXJCcdh2GFT0Fu-CL_Ihc76jKrMH3LXfZlMjtk,12289
585
591
  sglang/srt/managers/detokenizer_manager.py,sha256=SpLxTsSPKBZfD-ZMhJ5zpPPGuUb8PmcYgFSL9CsurU4,10696
586
- sglang/srt/managers/io_struct.py,sha256=LNLG7utMidF63eYWlWYxQ3G2GzOqaQzQc-9ld2yaycI,37859
592
+ sglang/srt/managers/io_struct.py,sha256=_wMPKcCynoq6-rDOkcKXFOUjSXUzIEmTHES8760jbdI,38058
587
593
  sglang/srt/managers/mm_utils.py,sha256=VrJiv0k_zDa83BsfVv8rmckgXKsvF0Wz88bCAb6R1as,29601
588
594
  sglang/srt/managers/multimodal_processor.py,sha256=mLnGcUHIvjruwoVhqiz4j7RMYMsZnvQMH8JnKCITDlY,1805
589
595
  sglang/srt/managers/schedule_batch.py,sha256=xmWSPAsllmPaWtKNrf-P8IX6fuQMTD-vZuk1y01vUjA,76289
590
596
  sglang/srt/managers/schedule_policy.py,sha256=BFYItMXtrsO5157tpFhYwLpZ8NSlzR4_gNZC1iE10D4,22400
591
- sglang/srt/managers/scheduler.py,sha256=rwI-qSIeVkzxkmVXPPtbK0mkQ5WY3myU3lV_AImFJ-Y,122502
597
+ sglang/srt/managers/scheduler.py,sha256=AJAbKjP6rBePpb4iVoD9X-99z6-cOaxUBS3OUlXA92U,123741
598
+ sglang/srt/managers/scheduler_input_blocker.py,sha256=zP8xU_UmU2H0AB6sEqvivDcDB1QDgTauNDYDIJ7Nez4,3683
592
599
  sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=Ei-9lyfrSQAlL9_ULuJLAhA7qpvUzjsBDzBBTDdnt4Q,31394
593
600
  sglang/srt/managers/session_controller.py,sha256=dzlMNZlo20FTSl64QqK7y7pElsdCy8ICOWWBPTBVwgs,6040
594
601
  sglang/srt/managers/template_manager.py,sha256=RrwRA2oqId_PMQ98qJQGwIxMroOxiorl2sGC9ARou_0,8543
595
- sglang/srt/managers/tokenizer_manager.py,sha256=yAgnR5eYeiid8gwJSBNQ1tkrYd5dq0RHXWn8nO_UdvY,76284
596
- sglang/srt/managers/tp_worker.py,sha256=D_NoiX69ry8cKRUYjX4nhtlPLdz1dipJEJekTouNG70,11363
602
+ sglang/srt/managers/tokenizer_manager.py,sha256=tq0tOSTJLZF8AE1iXzG-NrJyS8PHmmgQFXFlYGvZFUM,78025
603
+ sglang/srt/managers/tp_worker.py,sha256=PKNYIkInFkCITimwXxvXwDcHEwTgfQUfOZjXFvRKo5I,11617
597
604
  sglang/srt/managers/tp_worker_overlap_thread.py,sha256=kK0pn8mz-XgLGvVlJK6LvDVp4lnFk5vJH-Cs9NQ5Un4,10959
598
605
  sglang/srt/managers/utils.py,sha256=9_VGE48EK0PXVJ26aYvbRJ6n7gIZALvCcf6uZCccCgM,1369
599
606
  sglang/srt/mem_cache/allocator.py,sha256=id4riucsG9d-P1ikj3-DLJbzhM8-MuJfvDmB8H4g9-o,23364
@@ -611,7 +618,7 @@ sglang/srt/metrics/collector.py,sha256=FH6Wil2lMK7d7q74CfxfdULCaHKDXeAoo8UBZqqde
611
618
  sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
612
619
  sglang/srt/model_executor/cuda_graph_runner.py,sha256=6njRGysrhBDi2bAt5DyjGnyrexYvuEVsAJwQBBW8Ut0,31886
613
620
  sglang/srt/model_executor/forward_batch_info.py,sha256=NvUhReevoy6UfO4991P_5MPQyfefC3BOoTwTqUwMbYk,38430
614
- sglang/srt/model_executor/model_runner.py,sha256=UkI7i73Noaar9__sYVDGA-xEQHYSllZPUxRIbLwduWU,72373
621
+ sglang/srt/model_executor/model_runner.py,sha256=jC2Int_-_LQMYxQHfnXa76IMC7anMAbWpqgcsJweh28,72587
615
622
  sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
616
623
  sglang/srt/model_loader/loader.py,sha256=wxSUYrqv23w9yHYy9Lm0ozWggs9pf6Co4yqe1UO8PZk,59799
617
624
  sglang/srt/model_loader/utils.py,sha256=svSrPQxViBz_-0_sByu4coCYrzJ51-4FCi_bFqa6rDo,4474
@@ -625,7 +632,7 @@ sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,155
625
632
  sglang/srt/models/deepseek.py,sha256=sqobTr9a9c5pNLNJZE-WrJIE_qBq0lC5gCGv9NpGU_Q,17364
626
633
  sglang/srt/models/deepseek_janus_pro.py,sha256=td8xGs6ARfJ8AQCYwUhMOZoWigrAs7m3trF5-kXCqik,70418
627
634
  sglang/srt/models/deepseek_nextn.py,sha256=47fehxRdiOizr0rdLg5f1fzQEx6gGAOcDcWKtblloyk,5928
628
- sglang/srt/models/deepseek_v2.py,sha256=d6itbiEH_Zj0DdvH10dbxwTN7kK4g7rQYJVQ6A2-B70,105428
635
+ sglang/srt/models/deepseek_v2.py,sha256=S2KRJczyfTZ6Yl3PqF_2GAL7yUXHPl2BDhnm07JA_UE,103547
629
636
  sglang/srt/models/deepseek_vl2.py,sha256=7X3gI4DE7guUtNJvMLf06jO8MUHKa2Aan8evZg5hsXc,13061
630
637
  sglang/srt/models/exaone.py,sha256=TpO-rtCpEZ8Ua7hGFnS8l2oAYhY0Pij50grc9WQ2mvc,13576
631
638
  sglang/srt/models/gemma.py,sha256=2atKwFN5wCsJCmBRmURX9vaCcFjxlFirL7xsnG5KPlI,14180
@@ -693,7 +700,7 @@ sglang/srt/models/qwen2_moe.py,sha256=drTwlJlP0PLDl-Rl2h4GaOmOxexV1SMjMP97izhVpL
693
700
  sglang/srt/models/qwen2_rm.py,sha256=-mQXDEv11p-I1HXgYLTtY6ROem6UYorO958WsDrzsgs,2837
694
701
  sglang/srt/models/qwen2_vl.py,sha256=ZRAL32Ymt4ZCBqB0JorXAm4ZBrECZ9EbaXKaTSikkQM,21366
695
702
  sglang/srt/models/qwen3.py,sha256=COSC1TsCQNTq3E1QoVJYj1yTn-CFEVEX7keRwrrBXLs,20121
696
- sglang/srt/models/qwen3_moe.py,sha256=Vcdt77gDniDoZiAgsdhcWHtw1GiH-3Pk8GpiUxpjXWc,34019
703
+ sglang/srt/models/qwen3_moe.py,sha256=XJ-93F8gVWV5u8MsvC0DKwIDqSBgORO60l8Vd6T0dCE,31541
697
704
  sglang/srt/models/registry.py,sha256=bH9H0OiNV0Cn7FRBM30Oh6dxICTs9mJscdLtHYPJvxc,3783
698
705
  sglang/srt/models/roberta.py,sha256=3k53V2Gbezk3jU-D03thXx8csGn5DxFK01ZQ9WFiWPI,9828
699
706
  sglang/srt/models/siglip.py,sha256=DIVJRwdtpLD2QT1kVPIHw7Bn9BE40xJbkHujDNzkjgA,9892
@@ -763,13 +770,13 @@ sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg
763
770
  sglang/test/test_marlin_moe.py,sha256=PITMGcGF4ci2csXJejdLSNuxpURnFmcflsipAZsGakU,7991
764
771
  sglang/test/test_marlin_utils.py,sha256=gFMJwxbfEEJDeHI-HU5FDF9FFqqcM-yahK0TWyVIVgQ,5171
765
772
  sglang/test/test_programs.py,sha256=DJIbTChloPwaDiEWHQoSTuKv3BGx0-K8Y5e1mDDTACM,18910
766
- sglang/test/test_utils.py,sha256=hnT1lvTERIpmTapJKttFPfZCqHjNsTqiWNqpn5r-6CU,39454
773
+ sglang/test/test_utils.py,sha256=672Bp2kighQGrTC4SnIyy8rsoLOxGDjnCtsIhhOBLH4,41171
767
774
  sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
768
775
  sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
769
776
  sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
770
777
  sglang/test/attention/test_prefix_chunk_info.py,sha256=hpoDe2wfSa6RlUbfyri_c0iyBTb35UXGL9I2Xh6jamM,7772
771
- sglang-0.4.9.post5.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
772
- sglang-0.4.9.post5.dist-info/METADATA,sha256=jv_hoxV9ahzQ3tRZRibHpMxmYK6-lCJUDCd09LK4dPU,27348
773
- sglang-0.4.9.post5.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
774
- sglang-0.4.9.post5.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
775
- sglang-0.4.9.post5.dist-info/RECORD,,
778
+ sglang-0.4.9.post6.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
779
+ sglang-0.4.9.post6.dist-info/METADATA,sha256=CYETYraU2OqIt-j8SUxABCh-qIJAer5LaHRf0NZapvs,27394
780
+ sglang-0.4.9.post6.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
781
+ sglang-0.4.9.post6.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
782
+ sglang-0.4.9.post6.dist-info/RECORD,,