sglang 0.4.10__py3-none-any.whl → 0.4.10.post1__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (35) hide show
  1. sglang/bench_offline_throughput.py +20 -0
  2. sglang/srt/configs/model_config.py +1 -0
  3. sglang/srt/disaggregation/launch_lb.py +5 -20
  4. sglang/srt/disaggregation/mooncake/conn.py +33 -15
  5. sglang/srt/layers/attention/trtllm_mla_backend.py +372 -0
  6. sglang/srt/layers/attention/utils.py +6 -1
  7. sglang/srt/layers/moe/ep_moe/layer.py +19 -34
  8. sglang/srt/layers/moe/fused_moe_triton/layer.py +56 -2
  9. sglang/srt/layers/quantization/fp8.py +52 -0
  10. sglang/srt/layers/quantization/w8a8_int8.py +4 -1
  11. sglang/srt/managers/cache_controller.py +35 -35
  12. sglang/srt/managers/scheduler.py +1 -0
  13. sglang/srt/mem_cache/hicache_storage.py +15 -6
  14. sglang/srt/mem_cache/hiradix_cache.py +21 -4
  15. sglang/srt/mem_cache/memory_pool.py +15 -118
  16. sglang/srt/mem_cache/memory_pool_host.py +350 -33
  17. sglang/srt/mem_cache/nixl/hicache_nixl.py +163 -0
  18. sglang/srt/mem_cache/nixl/nixl_utils.py +238 -0
  19. sglang/srt/mem_cache/nixl/test_hicache_nixl_storage.py +216 -0
  20. sglang/srt/mem_cache/storage/hf3fs/client_hf3fs.py +8 -2
  21. sglang/srt/model_executor/cuda_graph_runner.py +25 -1
  22. sglang/srt/model_executor/model_runner.py +8 -1
  23. sglang/srt/model_loader/weight_utils.py +2 -0
  24. sglang/srt/models/deepseek_v2.py +5 -6
  25. sglang/srt/models/glm4_moe.py +3 -3
  26. sglang/srt/models/step3_vl.py +0 -3
  27. sglang/srt/server_args.py +40 -6
  28. sglang/srt/utils.py +1 -0
  29. sglang/test/attention/test_trtllm_mla_backend.py +945 -0
  30. sglang/version.py +1 -1
  31. {sglang-0.4.10.dist-info → sglang-0.4.10.post1.dist-info}/METADATA +1 -1
  32. {sglang-0.4.10.dist-info → sglang-0.4.10.post1.dist-info}/RECORD +35 -30
  33. {sglang-0.4.10.dist-info → sglang-0.4.10.post1.dist-info}/WHEEL +0 -0
  34. {sglang-0.4.10.dist-info → sglang-0.4.10.post1.dist-info}/licenses/LICENSE +0 -0
  35. {sglang-0.4.10.dist-info → sglang-0.4.10.post1.dist-info}/top_level.txt +0 -0
sglang/version.py CHANGED
@@ -1 +1 @@
1
- __version__ = "0.4.10"
1
+ __version__ = "0.4.10.post1"
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: sglang
3
- Version: 0.4.10
3
+ Version: 0.4.10.post1
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
@@ -1,6 +1,6 @@
1
1
  sglang/__init__.py,sha256=x1v8748WXS2BAptWAd5fR3YBEkngC7e_P_rkCNuNCGg,1702
2
2
  sglang/api.py,sha256=rcp3GeoyZhmJ0GDLPRkuZNcxd0TBJy_wfUDpcmQoqW8,7210
3
- sglang/bench_offline_throughput.py,sha256=TwgXZYmwPaHVsdPtNU9LO0p1tr5OOKLy9wYgrfGAlFU,14056
3
+ sglang/bench_offline_throughput.py,sha256=Bge0KG4Uiv9CHsxDnqKdXJmG_TcxM4blz3CmvLmDRvM,14815
4
4
  sglang/bench_one_batch.py,sha256=kSqUxn8kEib6GkEdSDZgAaY6P1JuFL2BadbfVuXJjQs,19610
5
5
  sglang/bench_one_batch_server.py,sha256=LS1BRmFwP67IpYlU5wmkWjqquiA5drvtWe-fwHnFSKI,14170
6
6
  sglang/bench_serving.py,sha256=nOjDnqOKDezDvtCvEhC1_FAoWVkir38bSaEIcSGGVZg,71748
@@ -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=C_r3OF-OA6DwUz6KYQF6xWUBApjHaqSPsV9JccafrWI,16403
13
- sglang/version.py,sha256=N_k8mdXQaZTz0YYxAgWi2g6nf_GP6B5r8Q49Om9EynA,23
13
+ sglang/version.py,sha256=UbzK-YnSSFDgJG6Hobrf-GLxjV7dEUOjLLPQZxNXD3E,29
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
@@ -40,10 +40,10 @@ sglang/srt/operations_strategy.py,sha256=Pwd2sKeRtKh9WJXgzlNr2tU9y6YMcI3MDLbatHq
40
40
  sglang/srt/patch_torch.py,sha256=OUPCGQSQz3MVZB1zZ_Eq8lXiw0uIKJ_HWjqQolI8FsM,3088
41
41
  sglang/srt/poll_based_barrier.py,sha256=9vjn32ROlSueEk7EuZjWKUrrL7XVR9KJsSVliGTYB_4,937
42
42
  sglang/srt/reasoning_parser.py,sha256=pN7SbLjaiyT5tdrAZgOFWy3QrDKnZEHkBFeIKFXknZM,9288
43
- sglang/srt/server_args.py,sha256=JXzikQ5RaVRlxV_zVQRmJUN7oCdgXGOiltFyMF0yX64,86103
43
+ sglang/srt/server_args.py,sha256=0L6rOyOSlGzScn3L_ifLNNLmzcYQZmKBfCL4AMqwQHw,87628
44
44
  sglang/srt/torch_memory_saver_adapter.py,sha256=K_eTx0UU84MHSTXI3iqYLdHV4IWtJMJ2FKdGFJR8v1E,2417
45
45
  sglang/srt/two_batch_overlap.py,sha256=AgaPeAXJQjt4RVhIVKR2YMEZRL8NqjN93NpgDE_fXfQ,28892
46
- sglang/srt/utils.py,sha256=6YkLoFlKyLpCNlk6zzaM5ys5_dMRTMOWNoLH09oEsOs,93321
46
+ sglang/srt/utils.py,sha256=qpwCJ9_JNgviPSiuCmX4n2FCNxkiAKFTqf1l2BRvmOo,93364
47
47
  sglang/srt/warmup.py,sha256=zldxhMlXpclRAJXmfBjJNUJd1eDizVdysibBvQyTVuA,1782
48
48
  sglang/srt/configs/__init__.py,sha256=gas3cKYVgZp7qF4m8FejX4EUUmdQd_w5WDTm7fo7yM0,743
49
49
  sglang/srt/configs/chatglm.py,sha256=j-b0YkdYUmQm2y1kNmMJtKeACxWKmBbvNNkDWbs6kbI,2907
@@ -56,7 +56,7 @@ sglang/srt/configs/janus_pro.py,sha256=yI2c_aWBIKiTcK6RZxK6hq8CIxyYRH_NN8uaSNKsr
56
56
  sglang/srt/configs/kimi_vl.py,sha256=4W7VQI3pr888ZsFA2SqCQo4mI0seXTOrGQ-x3oTvWew,1358
57
57
  sglang/srt/configs/kimi_vl_moonvit.py,sha256=hx2Rt4JSFbvy2HUTeLjBpge87m8M6ITAhqsgdNf_Jd4,1163
58
58
  sglang/srt/configs/load_config.py,sha256=qs-AxuplouBx2tsv9KGBOLZPbwzuVA4vbktbGP_cRp8,3309
59
- sglang/srt/configs/model_config.py,sha256=Qs0pZs_xcD8IavE7ZU_bz0p7R2VEjq9hvYzyJFmR3oQ,28837
59
+ sglang/srt/configs/model_config.py,sha256=opeGqkpJcCjPLNLSa_Wf2PfQPB17leJKYO_VgYi17q8,28888
60
60
  sglang/srt/configs/step3_vl.py,sha256=_Otgnym57DVgB_kZ__8c1_Ys5gSalA_K0ZuVjcG51T0,4845
61
61
  sglang/srt/configs/update_config.py,sha256=D-myV9Ombwp6wTF0RX9qUO6m5rxFiUcyse32FQADG28,4787
62
62
  sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
@@ -82,7 +82,7 @@ sglang/srt/debug_utils/text_comparator.py,sha256=pg58q3fzsnd3rJKRgZTdcOviOtAv8pS
82
82
  sglang/srt/disaggregation/decode.py,sha256=xPXKHz0e8O9GZetHiGVQoN8iIwEA1GwO8jj0NkCzgp4,33884
83
83
  sglang/srt/disaggregation/decode_schedule_batch_mixin.py,sha256=sUXmqiJGwuYoJvZjRCgTMVEN7bLefnEUlCTJpVA6ecI,6416
84
84
  sglang/srt/disaggregation/kv_events.py,sha256=yFbtwOeblcCmOXTjg00TopxnyjkmCBQIVz46KB8jetY,13555
85
- sglang/srt/disaggregation/launch_lb.py,sha256=mcbAztN4gnHevw_T5R2_nWsymsDEY9vHkm7OJ1vr6cc,4211
85
+ sglang/srt/disaggregation/launch_lb.py,sha256=Elw-J-97jTA70Ml1EMp_J-k9X9H5WHTWOC2XEvmyJRY,3827
86
86
  sglang/srt/disaggregation/mini_lb.py,sha256=bpnz6d90nv7vAoiPxmxFKlM1Gjct0R4IPkF9SwwBk9I,14504
87
87
  sglang/srt/disaggregation/prefill.py,sha256=6SN5df3PGWOixdEaTVR-INJjeSNXGP4X0y2IepsCZFs,24090
88
88
  sglang/srt/disaggregation/utils.py,sha256=ql5p9yjBCRcR0YMgcgRK6tH0kPGvLNTUgeXBr_BUcoI,12198
@@ -97,7 +97,7 @@ sglang/srt/disaggregation/common/utils.py,sha256=SxRhAWisNK8seGhb5BXBJ5u53DF7yeK
97
97
  sglang/srt/disaggregation/fake/__init__.py,sha256=jJGWdXwaQiGIoR6atKqkQfkJmVyQ09l55VUN2WjwaeY,77
98
98
  sglang/srt/disaggregation/fake/conn.py,sha256=oD1DArn1yDFZCu-X6p93uSLlAXEkt9lYxERICMznxGw,2286
99
99
  sglang/srt/disaggregation/mooncake/__init__.py,sha256=0TgqkAdQI1YynbHY6c0QISvVoOSk-0SwCIq5rjPSmgE,156
100
- sglang/srt/disaggregation/mooncake/conn.py,sha256=4TnmzcCDDm15j3XFFw_n_awq958-ljVahy8nC3CuFdc,61842
100
+ sglang/srt/disaggregation/mooncake/conn.py,sha256=BrOcdnqk7Ut90qDqajNzQx3qTBlnsVhLmMgUgXVGCdg,62677
101
101
  sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=-ulLhz6DBDsmEjwNjjH5Pj8ngZKISgwBc_zL3ViGDN0,4761
102
102
  sglang/srt/disaggregation/nixl/__init__.py,sha256=qODVPIGWUXKXq4zsRIcMYoAoAeg6nBIN9vdQOlVMANE,136
103
103
  sglang/srt/disaggregation/nixl/conn.py,sha256=eSof87fG21Dd4COszfnbeXIxne3TWvw0mSvCOkjsBZc,20323
@@ -188,7 +188,8 @@ sglang/srt/layers/attention/merge_state.py,sha256=OnWmNV6LYTUj0y8SCy_IK7A9q6v9bZ
188
188
  sglang/srt/layers/attention/tbo_backend.py,sha256=CBnSDUmDvT-Tbhi9NHNo1LLkgUAcIsWMJgKa-sAIKW8,9318
189
189
  sglang/srt/layers/attention/torch_native_backend.py,sha256=RsI4lYiQJAj-Xy0vKkzbyKOWHAVPOD9b0n0dEehwlos,9496
190
190
  sglang/srt/layers/attention/triton_backend.py,sha256=AryrPc4kzioxkiucPkApMA__0HoQTKOHNpwN0GEgm98,36923
191
- sglang/srt/layers/attention/utils.py,sha256=KiRSIHaGBhErUhPWCSD3Uiaqb0pSEF0NJFwTTQXoMks,2887
191
+ sglang/srt/layers/attention/trtllm_mla_backend.py,sha256=axNZ4VKRqbLvLFhmFjT-CESXjtGtu5sa6VypQdOXbhQ,13356
192
+ sglang/srt/layers/attention/utils.py,sha256=lx_jNw2VwlBe2VnDb1eXRRdN9WArw9rdWHTHqndUEFk,3174
192
193
  sglang/srt/layers/attention/vision.py,sha256=zI1Pq2Tx4SwOzYhIarxNtlTiZeVMWxVf8j-Pf7xhzCA,18111
193
194
  sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=eN0Lqpu5h27lMPuRnisX4jrU0WR061pCFSrh3vLlSD0,18936
194
195
  sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=FbgrPO5GvEGEvojV7Ss0jX7384e9vuzX2goDCQuP1Zk,31124
@@ -204,11 +205,11 @@ sglang/srt/layers/moe/router.py,sha256=UrPieRvemN7Ew48gtG7DA2xhNDBRSnZxzugTEBI-0
204
205
  sglang/srt/layers/moe/topk.py,sha256=ttPFvXzTY0eyh6UH8fJZCjyDp-R9EssSP8m229MxQ_4,26925
205
206
  sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
206
207
  sglang/srt/layers/moe/ep_moe/kernels.py,sha256=UFH-XacP3aBjAcn6_SeC641jGzjSrY0fz50IB9URix4,42706
207
- sglang/srt/layers/moe/ep_moe/layer.py,sha256=8vY_cIiHR5cNx0VSzHMj3AwtutgHOql3SexMrDU05dw,27613
208
+ sglang/srt/layers/moe/ep_moe/layer.py,sha256=zMhZy01cPePBSy4y_OLKT6XYxEZjtJ2Ec6CD69d2TXc,26911
208
209
  sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=CJhFTtOydsLSXMYNfzAASPI5p_oP7c4krGwX4bBcT1s,22909
209
210
  sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=4NXZHbCw-G-uSnNUj4up0yh3xBDPnT-x0pdoIr0lku8,831
210
211
  sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=Yb7LvdCQWonJC05_4UOMt3jQYWmRl-AkdY7H5EVwKhM,57575
211
- sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=aL3l9LUSD3fwNnezk1QRaz3mczfH_FtnVWJEdq81urk,25480
212
+ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=7DIbnX1QyVetaKh6ZTixTEedsL0QDmMd8_3pmFU9xnU,27716
212
213
  sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=HOEJJyehKTUIjkcsTc29aEXHqVLgDlv_jQL5fjvuUxU,5856
213
214
  "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
214
215
  "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
@@ -389,7 +390,7 @@ sglang/srt/layers/quantization/awq.py,sha256=gLEOx6QoGQ0McBEURXPVidKGsdNUjqc9sJw
389
390
  sglang/srt/layers/quantization/awq_triton.py,sha256=4XwJbRNlFmlvjbMGtWF1Wvk26Kf96162vNEmuI5Kars,11737
390
391
  sglang/srt/layers/quantization/base_config.py,sha256=CiSxG5iuOa-12V9kIOh0Fu8Y7knZxfFbbNGTo6nEaRE,7173
391
392
  sglang/srt/layers/quantization/blockwise_int8.py,sha256=3avetVUBjCiuYWBJQbElbVXWpx0SgsFsDc5F9F_zHgA,13879
392
- sglang/srt/layers/quantization/fp8.py,sha256=EILDyEkGQDyXRN_mmUd7z19vmOkAyNRDNsVHpmoPq4k,46422
393
+ sglang/srt/layers/quantization/fp8.py,sha256=DAateV96FCn2dWpgT_Zju2f0ARuAMOMd3HuNquXGk_E,48644
393
394
  sglang/srt/layers/quantization/fp8_kernel.py,sha256=8alryD9sEROh9Dh4cdPrGIpuOPbT6Hx6PNSe8fJc9zM,40856
394
395
  sglang/srt/layers/quantization/fp8_utils.py,sha256=6FVR6_Y8akP4ihCaGG_bLYurcC_PShPtiPgiEvGN5bE,25605
395
396
  sglang/srt/layers/quantization/gptq.py,sha256=3q9xfq5sJJHcrVr9C8-l9JdcLFvwKAAfWJEfZW6CzLs,39145
@@ -407,7 +408,7 @@ sglang/srt/layers/quantization/unquant.py,sha256=vgI2G1iHpMr2ys1-HaQ_YIXJN0z1m9u
407
408
  sglang/srt/layers/quantization/utils.py,sha256=9ZmfS0MMqueQ6zIxMIFoQy2OFBQ5vHjOHBgVvUn7ews,16213
408
409
  sglang/srt/layers/quantization/w4afp8.py,sha256=wa5t0g2IwVFcwHxZ213f5Vsnt8P4fum7WVoxvf0llpw,11092
409
410
  sglang/srt/layers/quantization/w8a8_fp8.py,sha256=I_W9A4Pnluaf-SaT8zq-tAJTt1acpxfE_kG5mqdc5cE,10250
410
- sglang/srt/layers/quantization/w8a8_int8.py,sha256=VGrVELArNuyw4FEtUefaeK8ayuX7cpXj1OOqCmOCzm4,34902
411
+ sglang/srt/layers/quantization/w8a8_int8.py,sha256=VXGAMKMlRU7k6c-WRXDExBTOa2oLODRFq-RIZXGsI-c,35013
411
412
  sglang/srt/layers/quantization/compressed_tensors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
412
413
  sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=DsPCI82A4rqI6mjKo-WorIKUweppFV7-16Qku-zW9_g,25897
413
414
  sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=vwcgwGuPhH2kHBfg7DnsndzffrbZx8xuFsl9Qeegd9s,24286
@@ -587,7 +588,7 @@ sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=QURCYxHNR8Ls4SQtt3dvdgjvdDVh
587
588
  sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=iz5scFNo2YFTeiM8beWg9Z1oZI-6AM_T1wBMCQ6qp2Q,6485
588
589
  sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=Ksova04wgeGsFqGOXWqJtMYaHgyUYcx8VU42BZQOkVA,5129
589
590
  sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=VqCAFvUtq_l-0RGIkx3W_fzD55QcW2FAcqpvSHOBFp4,5115
590
- sglang/srt/managers/cache_controller.py,sha256=_vilEi3xCkqazLiNhUu0fkNd1Cwc_Wlc85xwJZyotFg,29133
591
+ sglang/srt/managers/cache_controller.py,sha256=qQrYJk8DG7DDwdmAe-vOKUW36E_OEFP4rBCrKPXldOE,29265
591
592
  sglang/srt/managers/configure_logging.py,sha256=8sNXZ2z9pBWOwn-X3wyz013Ob8Nbm1zDxRkxoZjH-l4,1633
592
593
  sglang/srt/managers/data_parallel_controller.py,sha256=omM1s7os0rbMQSDlrCJcdLTWHx0eGxJ-DBVf9PX4kow,12412
593
594
  sglang/srt/managers/detokenizer_manager.py,sha256=SpLxTsSPKBZfD-ZMhJ5zpPPGuUb8PmcYgFSL9CsurU4,10696
@@ -596,7 +597,7 @@ sglang/srt/managers/mm_utils.py,sha256=VrJiv0k_zDa83BsfVv8rmckgXKsvF0Wz88bCAb6R1
596
597
  sglang/srt/managers/multimodal_processor.py,sha256=mLnGcUHIvjruwoVhqiz4j7RMYMsZnvQMH8JnKCITDlY,1805
597
598
  sglang/srt/managers/schedule_batch.py,sha256=xmWSPAsllmPaWtKNrf-P8IX6fuQMTD-vZuk1y01vUjA,76289
598
599
  sglang/srt/managers/schedule_policy.py,sha256=BFYItMXtrsO5157tpFhYwLpZ8NSlzR4_gNZC1iE10D4,22400
599
- sglang/srt/managers/scheduler.py,sha256=P-ALjdQxU3gyz5HXWXVewNnNlDKqDq-xcbJIXiXPJLg,100147
600
+ sglang/srt/managers/scheduler.py,sha256=JErd7exLcqADWzVV1lr2WSWKOpZfGUXuDhEBS9aWViw,100218
600
601
  sglang/srt/managers/scheduler_input_blocker.py,sha256=zP8xU_UmU2H0AB6sEqvivDcDB1QDgTauNDYDIJ7Nez4,3683
601
602
  sglang/srt/managers/scheduler_metrics_mixin.py,sha256=-1IOE30eOoTvvpz-Xg9J4UcamwAJle2-odFJ9vrinLE,9341
602
603
  sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=Ei-9lyfrSQAlL9_ULuJLAhA7qpvUzjsBDzBBTDdnt4Q,31394
@@ -612,27 +613,30 @@ sglang/srt/mem_cache/allocator.py,sha256=id4riucsG9d-P1ikj3-DLJbzhM8-MuJfvDmB8H4
612
613
  sglang/srt/mem_cache/base_prefix_cache.py,sha256=hLS2ncTMAz7Kpdk5pNwn5c6g8b61_K9OCBL4Vj1xsYc,2801
613
614
  sglang/srt/mem_cache/chunk_cache.py,sha256=FU_fq8e_dodgU8EkRqE4Jr0sKwCXQnKmapO88k6_JfU,3207
614
615
  sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
615
- sglang/srt/mem_cache/hicache_storage.py,sha256=dJK_WB1wYEh7ab6A-PJkLjDrCstArrG5OxcOGL3UfG8,6313
616
- sglang/srt/mem_cache/hiradix_cache.py,sha256=owr16BBs0TqmPd3LC9Dt5p1UPh9u4ISZ7O4b2i2-Ml4,25267
617
- sglang/srt/mem_cache/memory_pool.py,sha256=jQdWgOr6hpMLbGaEJc4immBgZG_MZbx8LyOv029ASVw,40932
618
- sglang/srt/mem_cache/memory_pool_host.py,sha256=cJGg-u0kTsjKFOwebpRsoo3G3HDtJheFAkz2Cn8wtaI,13150
616
+ sglang/srt/mem_cache/hicache_storage.py,sha256=t7A8eIRdSep_005M3_BKGMSUWgKWBQL8RoBRrpJu_fQ,6713
617
+ sglang/srt/mem_cache/hiradix_cache.py,sha256=QbKT2NKLuVe-1AqEc_njtZHXrq8d6WiNNbLa6Qio9qI,25818
618
+ sglang/srt/mem_cache/memory_pool.py,sha256=i-QhFrYuaAFcMOG7QH6SJyJ8hFzVRS5fjxwKmae8f94,37064
619
+ sglang/srt/mem_cache/memory_pool_host.py,sha256=M5dTu4BiEEiTlD7vtctd_LQrCODS1iradGaFtHaRkXM,25385
619
620
  sglang/srt/mem_cache/multimodal_cache.py,sha256=wZl2KeEl3xeoEsYdH33UoM-FO8kqfLo_XUgereJVvoM,1348
620
621
  sglang/srt/mem_cache/radix_cache.py,sha256=kp4JiVxh3eS4MsFxGKIYjYAb14B777qc1bHrABPqBMc,19101
621
622
  sglang/srt/mem_cache/swa_radix_cache.py,sha256=4qt_gS7GuHiekd6VvKkx54dB2GHJzNA_kj71Np6jTDA,40300
622
623
  sglang/srt/mem_cache/mooncake_store/mooncake_store.py,sha256=vFDsTE1a-6FP_AidlGKrlX9-oZvdxh8rnQiFDd-UcE8,9748
623
624
  sglang/srt/mem_cache/mooncake_store/unit_test.py,sha256=jMdJgU41VscKS-mn_AVZE8LQbb_3MWQ57i9V3OCWLI4,1007
624
- sglang/srt/mem_cache/storage/hf3fs/client_hf3fs.py,sha256=OZdY5Y4opSXi8zdLg92ZiHyNafLHWrGW-bMjoTqUdCc,5024
625
+ sglang/srt/mem_cache/nixl/hicache_nixl.py,sha256=GU2J2335razBIIEqV2L44CR8N-QYKmhbHEDGa61EFEE,6170
626
+ sglang/srt/mem_cache/nixl/nixl_utils.py,sha256=i2C70Ct2xdZtR_6OSL7uQk7ped1TieSpdUKo8yXdc08,8672
627
+ sglang/srt/mem_cache/nixl/test_hicache_nixl_storage.py,sha256=TBtN12muLFXZ296mK2d3TSTyj8qhMo5kAYWLpSjLBhY,8017
628
+ sglang/srt/mem_cache/storage/hf3fs/client_hf3fs.py,sha256=k9ZtqKfWdJh19jLyRu0iXJEciXXuqFcEuMXej3hJhmg,5178
625
629
  sglang/srt/mem_cache/storage/hf3fs/storage_hf3fs.py,sha256=H0X4v5AWs4uTHoBGgb9BgfKf3WZ3reZKpLNxmL1dTbI,9230
626
630
  sglang/srt/mem_cache/storage/hf3fs/test_hf3fs_utils.py,sha256=g2h0Woa2CK39pRTigHCVR_hq095dTaC70lCIAIs9THM,1089
627
631
  sglang/srt/metrics/collector.py,sha256=FH6Wil2lMK7d7q74CfxfdULCaHKDXeAoo8UBZqqdepM,20289
628
632
  sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
629
- sglang/srt/model_executor/cuda_graph_runner.py,sha256=6njRGysrhBDi2bAt5DyjGnyrexYvuEVsAJwQBBW8Ut0,31886
633
+ sglang/srt/model_executor/cuda_graph_runner.py,sha256=ViEnpEtqUeUOtd42WipMTrXWFaQ8jrkupEZ0UYrBKmE,32628
630
634
  sglang/srt/model_executor/forward_batch_info.py,sha256=NvUhReevoy6UfO4991P_5MPQyfefC3BOoTwTqUwMbYk,38430
631
- sglang/srt/model_executor/model_runner.py,sha256=oRYQpITQLF0xqdrJ-Jitk6uUUyd_41PjAi0_dU_v7fI,72778
635
+ sglang/srt/model_executor/model_runner.py,sha256=CtFvyzmUMX_-QMtytmBEWc6a7Q4SZsktOPjZO3ANGp4,73144
632
636
  sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
633
637
  sglang/srt/model_loader/loader.py,sha256=wxSUYrqv23w9yHYy9Lm0ozWggs9pf6Co4yqe1UO8PZk,59799
634
638
  sglang/srt/model_loader/utils.py,sha256=svSrPQxViBz_-0_sByu4coCYrzJ51-4FCi_bFqa6rDo,4474
635
- sglang/srt/model_loader/weight_utils.py,sha256=dfepjK1lhcFoxX6Vflb3imiHaj9_Tep_uZ7j3lfN9jE,38004
639
+ sglang/srt/model_loader/weight_utils.py,sha256=2UxbNYzoZxIIrDhsEOk6M-ofFg5s1UMtV1j23XRbnq4,38129
636
640
  sglang/srt/models/arcee.py,sha256=Q6fujRESEnljzs8ls3XGsPtWxzfKfUa0QAvAekYzeBY,18981
637
641
  sglang/srt/models/baichuan.py,sha256=HbvlErnkCSK4pRQYCSDxMcrn-1DQyfiNoeDcnRrJas8,15807
638
642
  sglang/srt/models/bert.py,sha256=ODJe8YfNRP-hHsomFWk4_QpcuiSsNfjzGf256EDS0Pc,15802
@@ -643,7 +647,7 @@ sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,155
643
647
  sglang/srt/models/deepseek.py,sha256=sqobTr9a9c5pNLNJZE-WrJIE_qBq0lC5gCGv9NpGU_Q,17364
644
648
  sglang/srt/models/deepseek_janus_pro.py,sha256=td8xGs6ARfJ8AQCYwUhMOZoWigrAs7m3trF5-kXCqik,70418
645
649
  sglang/srt/models/deepseek_nextn.py,sha256=47fehxRdiOizr0rdLg5f1fzQEx6gGAOcDcWKtblloyk,5928
646
- sglang/srt/models/deepseek_v2.py,sha256=9lnhzvt2ixxmV-uJSkSDu5Njj8juateAV68Ny_rU_QA,103661
650
+ sglang/srt/models/deepseek_v2.py,sha256=eHVdyXEiW7z0nIQByKt_h_Slfi6N-n9DWtKsBuBT9xE,103640
647
651
  sglang/srt/models/deepseek_vl2.py,sha256=7X3gI4DE7guUtNJvMLf06jO8MUHKa2Aan8evZg5hsXc,13061
648
652
  sglang/srt/models/exaone.py,sha256=TpO-rtCpEZ8Ua7hGFnS8l2oAYhY0Pij50grc9WQ2mvc,13576
649
653
  sglang/srt/models/gemma.py,sha256=2atKwFN5wCsJCmBRmURX9vaCcFjxlFirL7xsnG5KPlI,14180
@@ -655,7 +659,7 @@ sglang/srt/models/gemma3n_audio.py,sha256=isgKfjA5UieYawxU6medL2ssXlzYPqAbagDBnL
655
659
  sglang/srt/models/gemma3n_causal.py,sha256=kBDlpSWAg6ZDmUPe08Q7sj5CbdVD_0h07zwJw8sKu7I,36335
656
660
  sglang/srt/models/gemma3n_mm.py,sha256=pL_fvnkz9fKaC2Afms5KG3Me3u5sp1C2-mAUhqPGrQ4,18701
657
661
  sglang/srt/models/glm4.py,sha256=2VQzUqFkQTy_2nfkxP9SF6_9kKLTZUExGRjge7r99Es,11265
658
- sglang/srt/models/glm4_moe.py,sha256=Qt5IdWajKPJ_voG_yRlYikZtDvc3wskfGof8DktsUWo,39982
662
+ sglang/srt/models/glm4_moe.py,sha256=fV4gfrnhXPUTa4NnbzZha_euKpjbLi3JG7QGmJsiW4E,40007
659
663
  sglang/srt/models/glm4_moe_nextn.py,sha256=cWB6o3w7xBvgi2ouGKsqZ3iovKH65QiRQul42wcgzMk,5897
660
664
  sglang/srt/models/gpt2.py,sha256=kclhxEs8oJk1KCyhmAqo7rZqecVGGHYkc-a1WZi3aIk,9841
661
665
  sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtYhl4,10307
@@ -716,7 +720,7 @@ sglang/srt/models/registry.py,sha256=bH9H0OiNV0Cn7FRBM30Oh6dxICTs9mJscdLtHYPJvxc
716
720
  sglang/srt/models/roberta.py,sha256=3k53V2Gbezk3jU-D03thXx8csGn5DxFK01ZQ9WFiWPI,9828
717
721
  sglang/srt/models/siglip.py,sha256=DIVJRwdtpLD2QT1kVPIHw7Bn9BE40xJbkHujDNzkjgA,9892
718
722
  sglang/srt/models/stablelm.py,sha256=0x_31uIr3WcWwecdPAI3ek9KkyKBJS7VwknTk2y0gjY,12281
719
- sglang/srt/models/step3_vl.py,sha256=jKzj3I-Y8wi4D3CBRxy91UIKYDVKO2kBd_SiiEN8XuU,35242
723
+ sglang/srt/models/step3_vl.py,sha256=SAOpDqs0n53iYNbLHf1t9vjEw2GilQctAM71-KCcp2Q,35128
720
724
  sglang/srt/models/torch_native_llama.py,sha256=oTxxof5Wx2b2egIwWx56LeHjtPGQleQ2zM5jXHMiuOU,19293
721
725
  sglang/srt/models/transformers.py,sha256=_CL0npi-_HvqzxkLgFhv2EE3bfGvCLXPYAtPFywZOr0,10260
722
726
  sglang/srt/models/vila.py,sha256=3NiRHWOqXnCRjXmswAldsNtiE5zRILbScEsbhUzLZF4,9846
@@ -788,8 +792,9 @@ sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3h
788
792
  sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
789
793
  sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
790
794
  sglang/test/attention/test_prefix_chunk_info.py,sha256=hpoDe2wfSa6RlUbfyri_c0iyBTb35UXGL9I2Xh6jamM,7772
791
- sglang-0.4.10.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
792
- sglang-0.4.10.dist-info/METADATA,sha256=x8REsYLh39hvMvdAkoo3ZcLixFFwt8nyIMT63SOiKbA,27388
793
- sglang-0.4.10.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
794
- sglang-0.4.10.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
795
- sglang-0.4.10.dist-info/RECORD,,
795
+ sglang/test/attention/test_trtllm_mla_backend.py,sha256=D7KOJbsfGRSdO5dMlGokBpiC6Sh42g51HlVkgxSK-yI,36251
796
+ sglang-0.4.10.post1.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
797
+ sglang-0.4.10.post1.dist-info/METADATA,sha256=SWpzKC2hhUHjpZ9ZlfGfYDn8-P_qUeqwZU2kpZDnH_0,27394
798
+ sglang-0.4.10.post1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
799
+ sglang-0.4.10.post1.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
800
+ sglang-0.4.10.post1.dist-info/RECORD,,