sglang 0.4.5.post3__py3-none-any.whl → 0.4.6.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.
- sglang/bench_one_batch.py +19 -3
- sglang/bench_serving.py +8 -9
- sglang/compile_deep_gemm.py +45 -4
- sglang/srt/code_completion_parser.py +1 -1
- sglang/srt/configs/deepseekvl2.py +1 -1
- sglang/srt/configs/model_config.py +9 -3
- sglang/srt/constrained/llguidance_backend.py +78 -61
- sglang/srt/conversation.py +34 -1
- sglang/srt/disaggregation/decode.py +67 -13
- sglang/srt/disaggregation/fake/__init__.py +1 -0
- sglang/srt/disaggregation/fake/conn.py +88 -0
- sglang/srt/disaggregation/mini_lb.py +45 -8
- sglang/srt/disaggregation/mooncake/conn.py +198 -31
- sglang/srt/disaggregation/prefill.py +36 -12
- sglang/srt/disaggregation/utils.py +16 -2
- sglang/srt/entrypoints/engine.py +9 -0
- sglang/srt/entrypoints/http_server.py +35 -4
- sglang/srt/function_call_parser.py +77 -5
- sglang/srt/layers/attention/base_attn_backend.py +3 -0
- sglang/srt/layers/attention/cutlass_mla_backend.py +278 -0
- sglang/srt/layers/attention/flashattention_backend.py +28 -10
- sglang/srt/layers/attention/flashmla_backend.py +8 -11
- sglang/srt/layers/attention/utils.py +1 -1
- sglang/srt/layers/attention/vision.py +2 -0
- sglang/srt/layers/layernorm.py +38 -16
- sglang/srt/layers/logits_processor.py +2 -2
- sglang/srt/layers/moe/fused_moe_native.py +2 -4
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H20.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H200.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H20.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H200.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H20.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H200.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=96,device_name=NVIDIA_H20.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=264,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json +41 -41
- sglang/srt/layers/moe/fused_moe_triton/configs/E=272,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +20 -17
- sglang/srt/layers/moe/fused_moe_triton/layer.py +15 -17
- sglang/srt/layers/pooler.py +6 -0
- sglang/srt/layers/quantization/awq.py +5 -1
- sglang/srt/layers/quantization/deep_gemm.py +17 -10
- sglang/srt/layers/quantization/fp8.py +20 -22
- sglang/srt/layers/quantization/fp8_utils.py +2 -2
- sglang/srt/layers/quantization/int8_kernel.py +32 -1
- sglang/srt/layers/radix_attention.py +13 -3
- sglang/srt/layers/rotary_embedding.py +170 -126
- sglang/srt/managers/data_parallel_controller.py +10 -3
- sglang/srt/managers/io_struct.py +7 -0
- sglang/srt/managers/mm_utils.py +85 -28
- sglang/srt/managers/multimodal_processors/base_processor.py +14 -1
- sglang/srt/managers/multimodal_processors/deepseek_vl_v2.py +9 -2
- sglang/srt/managers/multimodal_processors/gemma3.py +2 -5
- sglang/srt/managers/multimodal_processors/janus_pro.py +2 -2
- sglang/srt/managers/multimodal_processors/minicpm.py +4 -3
- sglang/srt/managers/multimodal_processors/qwen_vl.py +38 -13
- sglang/srt/managers/schedule_batch.py +38 -12
- sglang/srt/managers/scheduler.py +41 -28
- sglang/srt/managers/scheduler_output_processor_mixin.py +25 -9
- sglang/srt/managers/tokenizer_manager.py +5 -1
- sglang/srt/managers/tp_worker.py +3 -3
- sglang/srt/managers/tp_worker_overlap_thread.py +9 -4
- sglang/srt/mem_cache/memory_pool.py +87 -0
- sglang/srt/model_executor/cuda_graph_runner.py +4 -3
- sglang/srt/model_executor/forward_batch_info.py +51 -95
- sglang/srt/model_executor/model_runner.py +19 -25
- sglang/srt/models/deepseek.py +12 -2
- sglang/srt/models/deepseek_nextn.py +101 -6
- sglang/srt/models/deepseek_v2.py +144 -70
- sglang/srt/models/deepseek_vl2.py +9 -4
- sglang/srt/models/gemma3_causal.py +1 -1
- sglang/srt/models/llama4.py +0 -1
- sglang/srt/models/minicpmo.py +5 -1
- sglang/srt/models/mllama4.py +2 -2
- sglang/srt/models/qwen2_5_vl.py +3 -6
- sglang/srt/models/qwen2_vl.py +3 -7
- sglang/srt/models/roberta.py +178 -0
- sglang/srt/openai_api/adapter.py +50 -11
- sglang/srt/openai_api/protocol.py +2 -0
- sglang/srt/reasoning_parser.py +25 -1
- sglang/srt/server_args.py +31 -24
- sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +3 -3
- sglang/srt/torch_memory_saver_adapter.py +10 -1
- sglang/srt/utils.py +5 -1
- sglang/test/runners.py +6 -13
- sglang/test/send_one.py +84 -28
- sglang/test/test_utils.py +74 -18
- sglang/version.py +1 -1
- {sglang-0.4.5.post3.dist-info → sglang-0.4.6.post1.dist-info}/METADATA +5 -6
- {sglang-0.4.5.post3.dist-info → sglang-0.4.6.post1.dist-info}/RECORD +97 -80
- {sglang-0.4.5.post3.dist-info → sglang-0.4.6.post1.dist-info}/WHEEL +1 -1
- {sglang-0.4.5.post3.dist-info → sglang-0.4.6.post1.dist-info}/licenses/LICENSE +0 -0
- {sglang-0.4.5.post3.dist-info → sglang-0.4.6.post1.dist-info}/top_level.txt +0 -0
@@ -1,16 +1,16 @@
|
|
1
1
|
sglang/__init__.py,sha256=T-fZEjKP66Q1q3PB56oREs5U3zf6bL0fNcdIbW8jMhE,1652
|
2
2
|
sglang/api.py,sha256=vHiKBg8wwIdmrpnGclop5BzJ-1Q88emrlrfLwNCHg98,7010
|
3
3
|
sglang/bench_offline_throughput.py,sha256=OQb-AjL4UNymmir02ht43uzgaNsnO_I11nXSowKMqBI,13841
|
4
|
-
sglang/bench_one_batch.py,sha256=
|
4
|
+
sglang/bench_one_batch.py,sha256=gUIYcFWM_oYSXnM4CHYJcyuX0l1aMG-afK7-iFjAJZI,19584
|
5
5
|
sglang/bench_one_batch_server.py,sha256=8VYNhaQbWGP8TkNVuy_sPjD5FiuVZHamtGRWKwa-Z-Q,5962
|
6
|
-
sglang/bench_serving.py,sha256=
|
6
|
+
sglang/bench_serving.py,sha256=8rbek9PLYEHdt8fdll-z_P9e6GpmlLohHiyqY99JXIs,57567
|
7
7
|
sglang/check_env.py,sha256=76itNLUw9KlqbiY1BI4u4YaMZaqyCNcrCLUIb6aHflM,8396
|
8
|
-
sglang/compile_deep_gemm.py,sha256=
|
8
|
+
sglang/compile_deep_gemm.py,sha256=Umy3oYFeCn40qHUdwlPVuFXmA24fFYB-fuWApgZnEfw,6211
|
9
9
|
sglang/global_config.py,sha256=xzLdk8W53fneFblNh8iIjGF9C3-7mnzR1-LleD9Btxg,1495
|
10
10
|
sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
|
11
11
|
sglang/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
|
12
12
|
sglang/utils.py,sha256=GIcgiRHkZ-gyPxXOdn1qFF41jkg4-YdDxbPc4mzO-qk,16159
|
13
|
-
sglang/version.py,sha256=
|
13
|
+
sglang/version.py,sha256=mb7cZWFtBTYPgotnX_1oAZadFITLHrAXwTSs2Eb1dvU,28
|
14
14
|
sglang/lang/chat_template.py,sha256=MwNL5dNTe8g_l2ljZubnrazEgT2xEv-9O2D0Ezwxy4I,19658
|
15
15
|
sglang/lang/choices.py,sha256=-W1DVw9N9ZliVpvmWrzIXG4cswAah8eMQrHWzkS3D8o,6234
|
16
16
|
sglang/lang/compiler.py,sha256=MAuzoOOpb98njJ7Io2SDmFkhTroDYiq0te0ZpfHkMY4,7597
|
@@ -26,28 +26,28 @@ sglang/lang/backend/runtime_endpoint.py,sha256=4pTG1TnOVTWBCh3rMhqL_NP2hsawG9xuP
|
|
26
26
|
sglang/lang/backend/vertexai.py,sha256=gz0uNYyBb88jbPYz6ZIJ774fefrcbuVdoK33bphUZpI,4827
|
27
27
|
sglang/srt/_custom_ops.py,sha256=L7NuEaRD_Q6Q54n0NZnLXgWZURbnn8Tkg4NQedE6zgA,3616
|
28
28
|
sglang/srt/aio_rwlock.py,sha256=6LYtOdeTUY3hkfa1dmYkgsaF2ttrwIF3hUWz2AZ2fqw,2970
|
29
|
-
sglang/srt/code_completion_parser.py,sha256=
|
30
|
-
sglang/srt/conversation.py,sha256=
|
29
|
+
sglang/srt/code_completion_parser.py,sha256=iYRFBxXBAysHcBnf9IHmmyjVkrqKu_9h6Z0_EEjjTp4,5404
|
30
|
+
sglang/srt/conversation.py,sha256=jgm15yl2SPjSlVjLPwWYklUsUUElq-7W6-KqqGc30vs,30262
|
31
31
|
sglang/srt/custom_op.py,sha256=J1PUcGaeJJjfAjp06BQsLpUkKyR1zsh9MvDiDlqqJsg,1129
|
32
|
-
sglang/srt/function_call_parser.py,sha256=
|
32
|
+
sglang/srt/function_call_parser.py,sha256=gkCzjf7F2xYUmRunrOKzuB_biTdTKxdA1Vil-v2NlCs,29546
|
33
33
|
sglang/srt/hf_transformers_utils.py,sha256=N2f-gA8yUq-UP_TJT276gNbDNzmddWsmWnq3px6TIj8,9342
|
34
34
|
sglang/srt/mm_utils.py,sha256=1ScBunw_x4W8ebM_AcJ62-1T2mfT8NlMJqdAhkF1lb0,12367
|
35
35
|
sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
|
36
36
|
sglang/srt/patch_torch.py,sha256=OUPCGQSQz3MVZB1zZ_Eq8lXiw0uIKJ_HWjqQolI8FsM,3088
|
37
|
-
sglang/srt/reasoning_parser.py,sha256=
|
38
|
-
sglang/srt/server_args.py,sha256=
|
39
|
-
sglang/srt/torch_memory_saver_adapter.py,sha256=
|
40
|
-
sglang/srt/utils.py,sha256=
|
37
|
+
sglang/srt/reasoning_parser.py,sha256=8AMk3XI8mfvz4AUuRHf_pNYpM_Mr64uT9EZ3o90cqQ8,6341
|
38
|
+
sglang/srt/server_args.py,sha256=jRHEskSyfmHbCnyqRzp3deI5HizDenDiyLjF65ZUqvg,55149
|
39
|
+
sglang/srt/torch_memory_saver_adapter.py,sha256=HYlgYJ2sgmjs2RSjU2KbCaXijRg3mTDZ0ZcCB5Bt6Ps,2211
|
40
|
+
sglang/srt/utils.py,sha256=u_YB-FXi3AY6mhRmk8wdPcKKAo1sZY0bMUgnjq8BtJI,63033
|
41
41
|
sglang/srt/warmup.py,sha256=FmJiYfjRr3X_eAe7ojQaPoN17LvHpjDmRWRnO-k86AQ,1469
|
42
42
|
sglang/srt/configs/__init__.py,sha256=vulncVn70WqIT6s0HaB8p_Q6FjOiaLwNZWpoJS9FIuQ,399
|
43
43
|
sglang/srt/configs/chatglm.py,sha256=j-b0YkdYUmQm2y1kNmMJtKeACxWKmBbvNNkDWbs6kbI,2907
|
44
44
|
sglang/srt/configs/dbrx.py,sha256=tdhIkXAQl1yr0MxqFmsDG1E0e2puRTTKm6UTyANBLac,11005
|
45
|
-
sglang/srt/configs/deepseekvl2.py,sha256=
|
45
|
+
sglang/srt/configs/deepseekvl2.py,sha256=21jZravchHcwyTQ5ROu1NzwI_eu-ngt3v8SRMm4XE0k,23327
|
46
46
|
sglang/srt/configs/device_config.py,sha256=kfmpPOECqYxcRoY-ko0QZRhyiBWUGP2CMF51DMUN5nU,435
|
47
47
|
sglang/srt/configs/exaone.py,sha256=Duxd4yQoKy8GWEzZD_kCY_OzmN_67CTJL_Kgn0eXk3g,10731
|
48
48
|
sglang/srt/configs/janus_pro.py,sha256=-QtJ4ZGZiAJb0AkOEcuCHzIKLw23nF8nRk3rdCcoUO0,19016
|
49
49
|
sglang/srt/configs/load_config.py,sha256=qs-AxuplouBx2tsv9KGBOLZPbwzuVA4vbktbGP_cRp8,3309
|
50
|
-
sglang/srt/configs/model_config.py,sha256=
|
50
|
+
sglang/srt/configs/model_config.py,sha256=GtVEAqxcitVldxLroaHYwoILjfa--a2KmbcBMyyeF08,22421
|
51
51
|
sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
|
52
52
|
sglang/srt/connector/__init__.py,sha256=czLX5JOxuMhH-T9eSJzoc1qv1B4z9chyffDRL5I6wo4,1247
|
53
53
|
sglang/srt/connector/base_connector.py,sha256=i6i1TIzsz4NbSEkrdMPq-urb2sN2aLAx8dazga4gB9U,2833
|
@@ -58,20 +58,22 @@ sglang/srt/connector/serde/__init__.py,sha256=MvBJ7FBJtnou-AD-EdsCdAIDCcA8kWvUNu
|
|
58
58
|
sglang/srt/connector/serde/safe_serde.py,sha256=Iv_mFsDvk-cXuw0WTykBZ2yGxW0jb82DwVqC7r3RmAU,750
|
59
59
|
sglang/srt/connector/serde/serde.py,sha256=n59I2MXLa7WCyN_8pEd8L-scJk7lMhmEX-GOUIhF0ZA,1004
|
60
60
|
sglang/srt/constrained/base_grammar_backend.py,sha256=ljTVWpBo3bolce-E_-mtHIY2XWez4qcyDPeaIeZyIhM,7454
|
61
|
-
sglang/srt/constrained/llguidance_backend.py,sha256=
|
61
|
+
sglang/srt/constrained/llguidance_backend.py,sha256=S3Mz6j1k816E0w0VE_iZwwtUa2uUOoEnleWtW9VXrbY,5952
|
62
62
|
sglang/srt/constrained/outlines_backend.py,sha256=XbmkZSJzJnnY7k11uj8Et3StfuOiFwRs3ID4IRYAA4Q,6839
|
63
63
|
sglang/srt/constrained/outlines_jump_forward.py,sha256=Gyubp-FVetxd6wP4FA_kD6cCXIRfr8k_ZDviJyte048,6824
|
64
64
|
sglang/srt/constrained/reasoner_grammar_backend.py,sha256=XFxdZqvPofmtCeIMqR10NOyph06HwbdXfiVI8rIoV5s,3646
|
65
65
|
sglang/srt/constrained/xgrammar_backend.py,sha256=oc3BTTe8mB5Szv_O-5nZzWbKEKFb22oUniqTUZhewLQ,7409
|
66
66
|
sglang/srt/constrained/triton_ops/bitmask_ops.py,sha256=WjTen9iuuFWLzkE1mAHQZB9_7aIy5QH8Wjf-lB-Fams,4614
|
67
|
-
sglang/srt/disaggregation/decode.py,sha256=
|
68
|
-
sglang/srt/disaggregation/mini_lb.py,sha256=
|
69
|
-
sglang/srt/disaggregation/prefill.py,sha256=
|
70
|
-
sglang/srt/disaggregation/utils.py,sha256=
|
67
|
+
sglang/srt/disaggregation/decode.py,sha256=nSHCBfEtD3a6c2a7XPAcCh4c0jw3BLG9EL-L3LlW_V0,25139
|
68
|
+
sglang/srt/disaggregation/mini_lb.py,sha256=zyJo20GI6G1ZIoBVY3ltcr0dDcH5qOJrtMfiGKGnBLI,10959
|
69
|
+
sglang/srt/disaggregation/prefill.py,sha256=4wLYQtPMbKWMQvF3mGnvr8ygd9xRLO9zTwLKeM5BIf8,15424
|
70
|
+
sglang/srt/disaggregation/utils.py,sha256=7gO734GOr4u03qwOf2UvFsfj4n-I37iyzQh7lFKbJRk,3501
|
71
71
|
sglang/srt/disaggregation/base/__init__.py,sha256=KR8xXoRCDAy2U623mfP6ujXu42m1_F9EiudjrKu2I_A,130
|
72
72
|
sglang/srt/disaggregation/base/conn.py,sha256=gpf32bhYXWm_iaYB6WcrDaJ-UoL1ZzPI_xpi5pMhRQo,2443
|
73
|
+
sglang/srt/disaggregation/fake/__init__.py,sha256=zmfeKYXjonRhfFOck1c_mP7Q4cW5G0f1RsTwRivKu0s,47
|
74
|
+
sglang/srt/disaggregation/fake/conn.py,sha256=DKEVBgmzV3CNzZ0-r7rFV4orue2iP_7apEtgn-fcTEA,2552
|
73
75
|
sglang/srt/disaggregation/mooncake/__init__.py,sha256=1vacEHmWjf7zgbMPzsXKB08FqNKNCquJdUiDlO41BOk,122
|
74
|
-
sglang/srt/disaggregation/mooncake/conn.py,sha256=
|
76
|
+
sglang/srt/disaggregation/mooncake/conn.py,sha256=DQ_PTxq_nZHFZ4LxHDhCIvQFPA1xUbaw1Sleyqhkq6U,28224
|
75
77
|
sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=MxDAB9ZetRF1pFS2LP3FVHPtQ1HjIt_SK3UMaYHZ94o,2604
|
76
78
|
sglang/srt/disaggregation/nixl/__init__.py,sha256=n9HjrRk36WUcZNeetGWOh2pSriLp7GNTq7YYX9K3EtY,85
|
77
79
|
sglang/srt/disaggregation/nixl/conn.py,sha256=ZFyKZQtGrTw7lNi9BYNlfY-1idHFzerTfvVNHG2Uj5c,22652
|
@@ -88,39 +90,40 @@ sglang/srt/distributed/device_communicators/pynccl_wrapper.py,sha256=LblisImY9d6
|
|
88
90
|
sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=bbruDIM1GgKIdB6gi71_I0mpB179I-qyvwKuSj1Kaic,20816
|
89
91
|
sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
|
90
92
|
sglang/srt/entrypoints/EngineBase.py,sha256=xoyvp6XAeDLY2_Q2Ng33H-fRhrXHv2ldJJKd-HuDhqE,1870
|
91
|
-
sglang/srt/entrypoints/engine.py,sha256
|
92
|
-
sglang/srt/entrypoints/http_server.py,sha256=
|
93
|
+
sglang/srt/entrypoints/engine.py,sha256=e5BPBIewPTQpYlk1c2eC4C_xyQgJ3mgNEm2Sg3GfV2s,22518
|
94
|
+
sglang/srt/entrypoints/http_server.py,sha256=vvyvCosUp5aTFlD8k4IyZDzj2yXQIsndhPkTl4u1nGc,29573
|
93
95
|
sglang/srt/entrypoints/http_server_engine.py,sha256=ihA6y3GXRs28Y9U3SgdQcJQjnw_SVIby7QrVgiafX04,4846
|
94
96
|
sglang/srt/entrypoints/verl_engine.py,sha256=XLYdwTwhH0jTjw8xczgZXWfBXMRb_ur2bg4TN0dTwfI,6975
|
95
97
|
sglang/srt/layers/activation.py,sha256=oSkdo8B8najXFcVay3Y__CEvgXh87lAIhG0CMp2Ugqs,5954
|
96
98
|
sglang/srt/layers/dp_attention.py,sha256=WJgXg_KyBzDHkwyfUFBowpDRFd0q5Q9LgEhqT-qT_ys,7549
|
97
99
|
sglang/srt/layers/elementwise.py,sha256=XCrR2i-9dP-H6jQo2zUuquwZrsl_wEQqj5Wxk6WUf7o,13987
|
98
|
-
sglang/srt/layers/layernorm.py,sha256=
|
100
|
+
sglang/srt/layers/layernorm.py,sha256=2XEaRK9e6syWO3YLcqWqlR7hZ5R-CFDqbCII-zntQLM,5957
|
99
101
|
sglang/srt/layers/linear.py,sha256=nC9MxJrFap1BEyqgFlBySH4IeQruIbcBp32cOhUl5Fw,52149
|
100
|
-
sglang/srt/layers/logits_processor.py,sha256=
|
102
|
+
sglang/srt/layers/logits_processor.py,sha256=RMcgimdBCOTg_ikyUvyqPJ_Bc_0rKBK5AtDTBPyQZAo,24230
|
101
103
|
sglang/srt/layers/parameter.py,sha256=zqWyEzpWzP4NNTjq3G9khq6XofgpcmJqQLg6Vd4WyWE,15084
|
102
|
-
sglang/srt/layers/pooler.py,sha256=
|
103
|
-
sglang/srt/layers/radix_attention.py,sha256=
|
104
|
-
sglang/srt/layers/rotary_embedding.py,sha256=
|
104
|
+
sglang/srt/layers/pooler.py,sha256=nR9F4x0GP2KYkJihhcL-TX30gfLY3mVhOLRfeBDeKL4,1942
|
105
|
+
sglang/srt/layers/radix_attention.py,sha256=F71GgDes_fEt_cHxR9HM2QhNG5u7uF4zDAuLqwLYEV4,3481
|
106
|
+
sglang/srt/layers/rotary_embedding.py,sha256=eVBwYvGpFhL1KyyPutQuZotmvSpChcxzyhpmcbQ6cKQ,48267
|
105
107
|
sglang/srt/layers/sampler.py,sha256=PNgMXm2vsNsE6Rt89R5GLDC3lDxdIujoWli8F3vldng,11384
|
106
108
|
sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSahyMfA,4063
|
107
109
|
sglang/srt/layers/vocab_parallel_embedding.py,sha256=QUxd4sELx6p3dHvEKmccPZ-phdd_9EjNdwjH3SJ9zxI,22238
|
108
|
-
sglang/srt/layers/attention/base_attn_backend.py,sha256=
|
110
|
+
sglang/srt/layers/attention/base_attn_backend.py,sha256=lGujcYJ_CxHJy0Q9L6Phn3ds-nBGMy0OGj3R54R65iQ,3334
|
111
|
+
sglang/srt/layers/attention/cutlass_mla_backend.py,sha256=kqtTVCIgDhcW5y9iWP8xcwGPuev-V5ipAUG-Ae3ot7g,9883
|
109
112
|
sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
|
110
|
-
sglang/srt/layers/attention/flashattention_backend.py,sha256=
|
113
|
+
sglang/srt/layers/attention/flashattention_backend.py,sha256=ysJt9pJ8pg_kVxvVUTvUL22-O7ABHCenLGGcqCotD6A,83206
|
111
114
|
sglang/srt/layers/attention/flashinfer_backend.py,sha256=YtMTgMhxxNrAbSoWTPJczgY4SR3WjnAPXPoJ2d5PUZY,46394
|
112
115
|
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=pnVhvVEK87iFW8gUb1G7X7c1tqro8R2DSEOFCnlV8Bo,30301
|
113
|
-
sglang/srt/layers/attention/flashmla_backend.py,sha256=
|
116
|
+
sglang/srt/layers/attention/flashmla_backend.py,sha256=IyE4w7GcNOxjjy3mQeuAMjPtBNvI-6JkoxvBlFxFvec,10270
|
114
117
|
sglang/srt/layers/attention/torch_native_backend.py,sha256=K5hUqBgakk2COSQqsaxWs0yEVOHS-7BlOygZTOeI8kE,9444
|
115
118
|
sglang/srt/layers/attention/triton_backend.py,sha256=oEEiUwHbm4rNw5ExbQ2c3n0TwAgkk77yuLFenj9bHOo,26902
|
116
|
-
sglang/srt/layers/attention/utils.py,sha256=
|
117
|
-
sglang/srt/layers/attention/vision.py,sha256=
|
119
|
+
sglang/srt/layers/attention/utils.py,sha256=rxB4sbNIHDTges78bDbnpd_hUgtyb3e16wUwgI4WmoU,2751
|
120
|
+
sglang/srt/layers/attention/vision.py,sha256=CtFU1wyz5191LcuyDzGJ01mB-mM-upPj2pXg4DO6wh4,11985
|
118
121
|
sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=DPu_aCPgwPqKWZPEQmp_xA7MPbpV2ip-MEICCB470Ao,19120
|
119
122
|
sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=vsDZZ5QGb8-KBzADgKshnVQbsW8zRJF1h5hgdPGW5lU,31124
|
120
123
|
sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=m12jEnQkNJguATqvZ57HtMC2hhU4wqdB8xAYdh25BxE,13922
|
121
124
|
sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=waZsmpKIp8rTgFSoM4QMabJuLaB3yW6ltOzAKJksBoE,6260
|
122
125
|
sglang/srt/layers/attention/triton_ops/rocm_mla_decode_rope.py,sha256=664WnAJ91EiCUZOcnVDfbTQf4uGJ4ZDZB1CbxpEUFZc,13866
|
123
|
-
sglang/srt/layers/moe/fused_moe_native.py,sha256=
|
126
|
+
sglang/srt/layers/moe/fused_moe_native.py,sha256=U0qh3udHuJJll3udydqABoXPFz0au9aEj8Lv7OAHYvQ,4655
|
124
127
|
sglang/srt/layers/moe/router.py,sha256=5Aeqoix_AS4uymb665OJE904wVSBkQeFdZP4e7KKPvg,10530
|
125
128
|
sglang/srt/layers/moe/topk.py,sha256=K-VU64nWBV07bu1Okn-uYbhz9gylq-KFNRYn2SFzu28,11129
|
126
129
|
sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
@@ -128,8 +131,8 @@ sglang/srt/layers/moe/ep_moe/kernels.py,sha256=ijqRzS-tb0LGnDU5hW-g0JH104ppADrWa
|
|
128
131
|
sglang/srt/layers/moe/ep_moe/layer.py,sha256=SZ0shPwgDp7xj-TCv9bfg5O7f2AXjF6xmBP5xkZ0Ips,36440
|
129
132
|
sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=zQV7Qr-Zrcr3D3efVvZepRQM02bj5djHPsijPssavk8,20430
|
130
133
|
sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=h9yMFAL_bagUf-qBED8gSWdCOb7d8IdA-pE-L_nIg8E,842
|
131
|
-
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=
|
132
|
-
sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=
|
134
|
+
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=13ygSeBoRkiqsERSHOIbIxLplVsSl-SUT6JxYPB-ViM,55968
|
135
|
+
sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=BMOV76fabrZcoyDmRpRbH11Jc0ogWH2k2QAQwvZIpgI,25084
|
133
136
|
"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
|
134
137
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=JJN0hryyLr5Zv3dSS7C8cPFhAwTT6XxUVnBGMZvV6JA,2752
|
135
138
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=ouRyZ5PEMPP2njPftCNhs-1g1y6wueWLmhI7G1SjV1k,4131
|
@@ -141,7 +144,19 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=XofcceHnkmqhVczEIVqGSbsxp
|
|
141
144
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=yf33YmWlVSjjyg0Q4OMAWvc9gjRxvttMrQBUEOfPl4I,4153
|
142
145
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=ZWMClYN1moVRUP2f0hYac38di_pUgZggyl9d2D5rnoc,4136
|
143
146
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=C65Q2Mv1LxFQ_qDnv11IZ9nwl7sGZo72nWDflMttu4g,4147
|
147
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=I3k416HbXU_rYb8scD8gAI4fuBlElHl06PM347Qa11w,3253
|
148
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H20.json",sha256=RgV8C4F1LO09h01YsgF_eqX6GNoBtC7ulPfJRUUbg_g,3241
|
149
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H200.json",sha256=nsNEuDNks0tVLfQfIm7xxFwEeptTfQcoa9fJy0NS8xQ,3247
|
150
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=qbqjisJ4oKmcYzumHPRk5UyOzsdi8J6xas82UWHMeAI,3263
|
151
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H20.json",sha256=vS2DRIDOqWyiBvbG6H746ownfkD1F8Aj2YZ0ET9xll8,3232
|
152
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=1n5XyZZ5sKAi-Z1duWOhLUfr6gkvnOpvxfbqIT6iU_4,3265
|
153
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H200.json",sha256=xqhl748it8GV2KXX0XixitE_ywnsKksqK8AGL7tAgT8,3254
|
144
154
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=512,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=FsWbV4Q6AzAtgegVuENBDz2ZcSJsqNiwUIVfQbpP7hQ,3244
|
155
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=IuvyC8TNhCVAmUZfLSoETsyCKsmejKXrs_0zuwFLPAU,3265
|
156
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H20.json",sha256=10Ntu2aVD5vGLonx-jW0qNw-tgZWdZmzMGx7utDVeng,3237
|
157
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=pdQ1RvXvdWDn8Y8-8MAX3vn-T-wbtkZvHV9GZZvNjnc,3266
|
158
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H200.json",sha256=JraM-Nvbg5V_TJkSl6UPFYZN1zHHoIbr2pAcksenoTY,3248
|
159
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=96,device_name=NVIDIA_H20.json",sha256=JtcHRlPz8xQEAqJ9EWI63oYvdmjQFG6VTHqtt85VOSA,3221
|
145
160
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=144,N=512,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=pk6VZChh2Y0CsJSzjtUhOnlta1QLTUEWy33aKQU47XY,3244
|
146
161
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=Gmk24hc5lVIfQtqSa5wLOcWKedMN8aZUe93DBh6J1AY,3249
|
147
162
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1024,device_name=NVIDIA_H200.json",sha256=uY_XMPomaXMXxIkTR4ctU_Ybri_jMv2VvCcV-f6O_bw,3255
|
@@ -187,8 +202,9 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=XofcceHnkmqhVczEIVqGSbsxp
|
|
187
202
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=64,device_name=NVIDIA_L40S,dtype=int8_w8a8.json",sha256=Sc9xK1wtRUqIzXppbutcq-Y2e9M0DZl2OGVzzB0aQuI,3265
|
188
203
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=264,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json",sha256=7YmtaXKnmX8DdYnUJ7WQFa7xjr2Yun9WIdQNoCf_K28,3255
|
189
204
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=264,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=3Zt4hbC3yJxWvP0T7K93YAPaUP8fQ1P1Wk0CGqtBga8,3259
|
190
|
-
"sglang/srt/layers/moe/fused_moe_triton/configs/E=264,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=
|
205
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=264,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=P8GpVR8fjrX7OFbBBFE4y4MJ4uhgoyUV4NYCm1qhWxk,3266
|
191
206
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=272,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json",sha256=4B0SmzRQ2-PsBJcFe7neM1OKfWpsbiY4x6c6COQNMsQ,3254
|
207
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=272,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=f5HTi_9fWvInEyJp8pFgaVN6A9vxu3_845eSZGN9Ypo,3264
|
192
208
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=272,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=Piw4LN6d8QYrUahWsw3XUOtTMD1o3vHPwA94sGI56Gk,3242
|
193
209
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=288,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=3T8_rF2PEojhgTMyQ8DscXgJCWWdWfDPj4M434zWcA4,3243
|
194
210
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=64,N=1024,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=TBscr1uWwpY0FrKQ5Y3EO_Qg6I97u4f_zjnWRvoeLvE,3260
|
@@ -265,15 +281,15 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=XofcceHnkmqhVczEIVqGSbsxp
|
|
265
281
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=8,N=8192,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json",sha256=sY2nWMPh9lsIkhPCjkHO245wpnfFbrHmzdcZDVFPVww,3265
|
266
282
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=8,N=8192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json",sha256=Uz5X80VcNBOaxshwVNUEittHk2zqB4HQCfTJ4TPG5aM,3274
|
267
283
|
sglang/srt/layers/quantization/__init__.py,sha256=UOQcyCvKFkX0u_OPPex7X5X98iUR3lXgBnLbffu0n9g,12424
|
268
|
-
sglang/srt/layers/quantization/awq.py,sha256=
|
284
|
+
sglang/srt/layers/quantization/awq.py,sha256=KemDG55U3B6YZVjMV71awVAIj0islFvtxcUHmOBeGy0,6739
|
269
285
|
sglang/srt/layers/quantization/base_config.py,sha256=jWk_egQrVNMYmQgbTI9vkcgzScLFjB5_sywFlAfE5J0,4776
|
270
286
|
sglang/srt/layers/quantization/blockwise_int8.py,sha256=cu9-JiCZDfMfvB97Kv_-eEG87VX5bRFIllFkzpO_xIg,15122
|
271
|
-
sglang/srt/layers/quantization/deep_gemm.py,sha256=
|
272
|
-
sglang/srt/layers/quantization/fp8.py,sha256=
|
287
|
+
sglang/srt/layers/quantization/deep_gemm.py,sha256=UFzsd0iiqVTBo0Ow_6ylVVFK8B9EUWTNQQYGvsNfm2s,13129
|
288
|
+
sglang/srt/layers/quantization/fp8.py,sha256=da-6ji_HBISKwIgMMX-JGlDKMLi-qL9j2XLer5cFAsU,40945
|
273
289
|
sglang/srt/layers/quantization/fp8_kernel.py,sha256=C2_hOLRO27-Yvjy-Nm2niehD2gWSMuP6TnNX07ESqh4,32018
|
274
|
-
sglang/srt/layers/quantization/fp8_utils.py,sha256=
|
290
|
+
sglang/srt/layers/quantization/fp8_utils.py,sha256=vqH-bMb2DD0A7Y7hZjN-TGTg5h6aJ-cLW9H2adyZzqk,18651
|
275
291
|
sglang/srt/layers/quantization/gptq.py,sha256=gyGMOPXHzozK7pPWSjKgLdFX9h7MCEww7n8FqEVEVac,15364
|
276
|
-
sglang/srt/layers/quantization/int8_kernel.py,sha256=
|
292
|
+
sglang/srt/layers/quantization/int8_kernel.py,sha256=CR-VuTTR4GYluOZTpS5mmEz3hYrsY4GOX-G-h3XAYKc,12163
|
277
293
|
sglang/srt/layers/quantization/int8_utils.py,sha256=YK9CS-lb_n91kNCTKK5o5apYF31V2giDg5G5VKrpcUA,2356
|
278
294
|
sglang/srt/layers/quantization/kv_cache.py,sha256=-yaFTdB75T0BbvQeuIpH6rZoL3R8t6OIJVGB-xdtpCw,3492
|
279
295
|
sglang/srt/layers/quantization/modelopt_quant.py,sha256=TpPgtbV7O5r1JY4Wm0np2pReQO6XERIdEDQcV41oTn0,16596
|
@@ -457,43 +473,43 @@ sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=ZmWEqHJaorRNNj-c_ZXPi_pX8X_yIA
|
|
457
473
|
sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=Q58UzWUb3QFqY_ZxWA3poN373N0Hwkks5AQRKIuvFC8,4517
|
458
474
|
sglang/srt/managers/cache_controller.py,sha256=d4RGqbut1FlzJnpqr7WY_TYmRjYPS07OoOVbztjs5xI,18959
|
459
475
|
sglang/srt/managers/configure_logging.py,sha256=fOJaXAQ1n9m-8KPJndpsKvS885i69SMafoEADLIVfIM,1633
|
460
|
-
sglang/srt/managers/data_parallel_controller.py,sha256=
|
476
|
+
sglang/srt/managers/data_parallel_controller.py,sha256=Oo-0sbF0W1fcpw88-iKH_7pttYjWl8IHCePcuF3rU5c,10894
|
461
477
|
sglang/srt/managers/detokenizer_manager.py,sha256=3S3aRvKSi75RQSxEEQkeyxKDNNunWiw9wlwsbT1VXSo,10099
|
462
478
|
sglang/srt/managers/expert_distribution.py,sha256=r3o5RGI0gnV7xb60AApqKYa0oiSB37oB7hQBX7P3xZM,3225
|
463
|
-
sglang/srt/managers/io_struct.py,sha256=
|
464
|
-
sglang/srt/managers/mm_utils.py,sha256=
|
479
|
+
sglang/srt/managers/io_struct.py,sha256=9mdBGOkblguT1x6Ds9wL3j0MWAQiUQVdVRL4a7IUnA4,31631
|
480
|
+
sglang/srt/managers/mm_utils.py,sha256=JTu5B7jZWTtZi8LCpVa6ITvSToxcuf5PDbb3FJC9M6o,18089
|
465
481
|
sglang/srt/managers/multimodal_processor.py,sha256=XlRYvNhF6XOssreRX9DZPhLSpps_VE62gSKw3EGdNPo,2088
|
466
|
-
sglang/srt/managers/schedule_batch.py,sha256=
|
482
|
+
sglang/srt/managers/schedule_batch.py,sha256=zUQGVjLbi9TK5tfyzHNMSAnPeNeFi9GFI2AC8Fr2pbo,63824
|
467
483
|
sglang/srt/managers/schedule_policy.py,sha256=E1qVq2G3jptKdX9nlqfayeRBUll9xB6bK8nBf3EW32E,19469
|
468
|
-
sglang/srt/managers/scheduler.py,sha256=
|
469
|
-
sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=
|
484
|
+
sglang/srt/managers/scheduler.py,sha256=7o03npmnu775d1DRDAkTJjl8OuJlE_xuM3BQji6BYLI,80808
|
485
|
+
sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=GxdkTR24_P_2C3ib0dc7Xqklrz8SiHtUTlM0c7AlKlk,26754
|
470
486
|
sglang/srt/managers/session_controller.py,sha256=o-ifit0n4_xHLNmyD0Ams8FxGRgxFybX-Vz1hwgr3UQ,5755
|
471
|
-
sglang/srt/managers/tokenizer_manager.py,sha256=
|
472
|
-
sglang/srt/managers/tp_worker.py,sha256=
|
473
|
-
sglang/srt/managers/tp_worker_overlap_thread.py,sha256=
|
487
|
+
sglang/srt/managers/tokenizer_manager.py,sha256=4l4PAvfQrJqlYADQbl7cgpLhBBY52pzI5AzRYIzAjLs,50693
|
488
|
+
sglang/srt/managers/tp_worker.py,sha256=LhbhCovDvab6Cx4faR88s4q_3D-Di9s5sKCndsDxF9E,8966
|
489
|
+
sglang/srt/managers/tp_worker_overlap_thread.py,sha256=lkG_yN6_UEv5mhmZ7cKP7_A5sIVMQw1GPwkqM91EWSE,9304
|
474
490
|
sglang/srt/managers/utils.py,sha256=5i75uLlQOF_5CaT02CrWtwozMTtwTg2_nLP8Dtr-JZQ,1536
|
475
|
-
sglang/srt/managers/multimodal_processors/base_processor.py,sha256=
|
491
|
+
sglang/srt/managers/multimodal_processors/base_processor.py,sha256=ata9H6Ry4QfqBoA_g0auG0sMnKfGrlZn74lM77ihtiA,10172
|
476
492
|
sglang/srt/managers/multimodal_processors/clip.py,sha256=lRc2mcuDbAhZVf-0EfkO81pqDiol9zLvTpDqtPIBQ2k,1525
|
477
|
-
sglang/srt/managers/multimodal_processors/deepseek_vl_v2.py,sha256=
|
478
|
-
sglang/srt/managers/multimodal_processors/gemma3.py,sha256=
|
479
|
-
sglang/srt/managers/multimodal_processors/janus_pro.py,sha256=
|
493
|
+
sglang/srt/managers/multimodal_processors/deepseek_vl_v2.py,sha256=hpjpGFzlRBQ8Xv08i37X_VUhnDp_Qm2xD1_F17vK8fI,3253
|
494
|
+
sglang/srt/managers/multimodal_processors/gemma3.py,sha256=G52ck_3UQGeyrtvjLqI8B0Tm8iNsyB_ahiMTAvx083U,2191
|
495
|
+
sglang/srt/managers/multimodal_processors/janus_pro.py,sha256=UJoKQWsoU9kittKDwjWbG2KC12wSA-4A3DpTPhA6VoI,1854
|
480
496
|
sglang/srt/managers/multimodal_processors/llava.py,sha256=8mac3vUUpVd12o43k1TyMaLEySZB915ks8Q5epeZmbg,6209
|
481
|
-
sglang/srt/managers/multimodal_processors/minicpm.py,sha256=
|
497
|
+
sglang/srt/managers/multimodal_processors/minicpm.py,sha256=uEnlsImjHBOMVNGlfBGpn1zCDLNeMY58HvJ7ZthL2N4,5698
|
482
498
|
sglang/srt/managers/multimodal_processors/mlama.py,sha256=MLiGS606LzVtdoXvjWGANx-K_7nE9J_fMVmkXN7Gz8k,1661
|
483
499
|
sglang/srt/managers/multimodal_processors/mllama4.py,sha256=50Yox7TaGrrB7iPjN1dQ_UzuY41x7VLmMcRXBhTgUvE,5592
|
484
|
-
sglang/srt/managers/multimodal_processors/qwen_vl.py,sha256=
|
500
|
+
sglang/srt/managers/multimodal_processors/qwen_vl.py,sha256=l94DOaY9vhlD-QjWVWNHUmLu48UKTb-QN9vXqrQxBgA,6907
|
485
501
|
sglang/srt/mem_cache/base_prefix_cache.py,sha256=NY62Zo0A0tLJ7ObRLOQqQcXCxoJUDZsK8f5U4dNQjKc,973
|
486
502
|
sglang/srt/mem_cache/chunk_cache.py,sha256=it5SfL1FwMbrdeOH-I-Eu_i-I9hFB1xL-z_brIUoCkk,1835
|
487
503
|
sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
|
488
504
|
sglang/srt/mem_cache/hiradix_cache.py,sha256=BJR-R2u5YyYIhGIxTY-3rf8Vx60XjCRU8Yhmkn2fzzM,16597
|
489
|
-
sglang/srt/mem_cache/memory_pool.py,sha256=
|
505
|
+
sglang/srt/mem_cache/memory_pool.py,sha256=J2eAAefAl0TIejH7h-hwz_ak_T-fSh_e45tUNrhX0BE,34599
|
490
506
|
sglang/srt/mem_cache/paged_allocator.py,sha256=BrJS0vN1k-vTSgb_M8u_1KoZFRgzgR1WRyImCTq3T0U,9770
|
491
507
|
sglang/srt/mem_cache/radix_cache.py,sha256=Lm-pco6CJ4orb9IfDpbHm5MnyK8Ya0OF1x9p88dv548,14906
|
492
508
|
sglang/srt/metrics/collector.py,sha256=zHg4twFQJvuK1mSme3-EYQa9PJryfp_u7a4RxQ5RcO0,8874
|
493
509
|
sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
|
494
|
-
sglang/srt/model_executor/cuda_graph_runner.py,sha256=
|
495
|
-
sglang/srt/model_executor/forward_batch_info.py,sha256=
|
496
|
-
sglang/srt/model_executor/model_runner.py,sha256=
|
510
|
+
sglang/srt/model_executor/cuda_graph_runner.py,sha256=iFryO9dglpnFCoNWxZqKdUhQycT8In29C0kIba3G1Dw,23687
|
511
|
+
sglang/srt/model_executor/forward_batch_info.py,sha256=T9B5vWaJwlKUH0fQTPe3XdbkTYEUI6iKxBxUHs-cAMM,26632
|
512
|
+
sglang/srt/model_executor/model_runner.py,sha256=O4vKZ4c-u69ZeKPBjAfiunvtnQHskZVmbUUK4fKFb5E,46417
|
497
513
|
sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
|
498
514
|
sglang/srt/model_loader/loader.py,sha256=YYmtvkQw0B1qgPw0_gN-K4yy7CEYbTSR__0Dl1Fnm6k,55342
|
499
515
|
sglang/srt/model_loader/utils.py,sha256=0NaMR67fESFopaklmsleiL27XH1QUrjZW246MUu1EJ0,1369
|
@@ -504,16 +520,16 @@ sglang/srt/models/chatglm.py,sha256=cajLN9caBl09e0TwOFkiTTKDqwlbmHo_yS-NCjdeQW8,
|
|
504
520
|
sglang/srt/models/clip.py,sha256=fCMtAcaKjruSIWfD4YGb4HXh6Tzp2pjpgDmp5JpwBPU,19794
|
505
521
|
sglang/srt/models/commandr.py,sha256=5Y_b3K0QY7D37nFGkyiGgY38RleRui_GJUYcHSuHUZo,15315
|
506
522
|
sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,15598
|
507
|
-
sglang/srt/models/deepseek.py,sha256=
|
523
|
+
sglang/srt/models/deepseek.py,sha256=ZnN02HdgXCB23Vno5V9UMUoOxH5HC82vNTwsVulUJ-o,17206
|
508
524
|
sglang/srt/models/deepseek_janus_pro.py,sha256=8wAzvcGdyo--3faMN4QtagT1eAZMhMFduvpCXqUS48Q,70456
|
509
|
-
sglang/srt/models/deepseek_nextn.py,sha256=
|
510
|
-
sglang/srt/models/deepseek_v2.py,sha256=
|
511
|
-
sglang/srt/models/deepseek_vl2.py,sha256=
|
525
|
+
sglang/srt/models/deepseek_nextn.py,sha256=XW0PJAvUVx5i1F6liNMooopj833qyQ4Y4ujn3iJDDak,17825
|
526
|
+
sglang/srt/models/deepseek_v2.py,sha256=6fEihiaHcl9tjawa1GnCKGIappuLnDfmmVChhPswSIU,71820
|
527
|
+
sglang/srt/models/deepseek_vl2.py,sha256=j8BdxZsMjm6lPdbDipEIKhVIVywCP1Vl1Kl46BZ5_0Y,13147
|
512
528
|
sglang/srt/models/exaone.py,sha256=rX7J0xFt9TSt6tMIhnYMkb5KDnqTJIV4BtjPLFwQ8_8,13425
|
513
529
|
sglang/srt/models/gemma.py,sha256=4cdrPISg1VKnsuI-QPTpYvet4BrX8BMKvCIN82iLskw,12641
|
514
530
|
sglang/srt/models/gemma2.py,sha256=kqtwdo93GWKm2iBN29RoIRH2ggRm-K_80LM5btgfBLo,16395
|
515
531
|
sglang/srt/models/gemma2_reward.py,sha256=V8U3_ADUHWPdOwvEe1jhGW-oJmBgL8t1TY3-67Ksv2A,2618
|
516
|
-
sglang/srt/models/gemma3_causal.py,sha256=
|
532
|
+
sglang/srt/models/gemma3_causal.py,sha256=LfwHhF0nRD7OnmeHXXfQ7rofnFXjJI74gZiptak18RY,24924
|
517
533
|
sglang/srt/models/gemma3_mm.py,sha256=tWX2vIdRf5zePwKMLbb0d24DUWoTdjmdXnxIcULQJ2E,15221
|
518
534
|
sglang/srt/models/gpt2.py,sha256=kclhxEs8oJk1KCyhmAqo7rZqecVGGHYkc-a1WZi3aIk,9841
|
519
535
|
sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtYhl4,10307
|
@@ -522,7 +538,7 @@ sglang/srt/models/grok.py,sha256=vESZeGS4adI_JAerXIkCcTm15-CNiGeS7VHc36C6w1A,280
|
|
522
538
|
sglang/srt/models/internlm2.py,sha256=RDAT9drjdgVEFmCMq99RTn3weMQFhl1NHhkhyDX8f7M,13056
|
523
539
|
sglang/srt/models/internlm2_reward.py,sha256=ndfGmyqYZbVZ7C7rJ-v9oK3wa-EpoBGybS8MlyKZi2E,2522
|
524
540
|
sglang/srt/models/llama.py,sha256=Y4ROe8ohP84G4vin_Sr_vjG0XRoM5gGgnrojxOXn_uc,24942
|
525
|
-
sglang/srt/models/llama4.py,sha256=
|
541
|
+
sglang/srt/models/llama4.py,sha256=idwV_rEJ_tPMD1iLQzvaJqmux-Osoa3bc0g04VYgu1w,17867
|
526
542
|
sglang/srt/models/llama_classification.py,sha256=4QWTFaUZIFKYZvEzs8bx8VkOZNIwdYCLrnwrdAw4QK0,3108
|
527
543
|
sglang/srt/models/llama_eagle.py,sha256=OB2lKsjn7BcfCZljklnhk83me8j0PuQmYLou7baNcq4,4866
|
528
544
|
sglang/srt/models/llama_eagle3.py,sha256=v3bftBVDIGjnzngQYnu19cy0J_3w7yruHqLP5nsAQDM,6642
|
@@ -532,35 +548,36 @@ sglang/srt/models/llava.py,sha256=KMwNNrlMuMaKEOZMDRBKBQbe6uctpKTLc0zOceyGC34,27
|
|
532
548
|
sglang/srt/models/llavavid.py,sha256=q0lHlRnoYHKJZsWnkIQdd6dYAQ26t7XsmrqA0zDGmZc,12829
|
533
549
|
sglang/srt/models/minicpm.py,sha256=m5HFsSJj0Po09LY9R6qj6K4gceqWDMOePz3NDGgMGT4,14691
|
534
550
|
sglang/srt/models/minicpm3.py,sha256=6-gfHSfXwyB5zw7AIj-c5TzjpEKR6CM0H67MrD8LVUE,19347
|
535
|
-
sglang/srt/models/minicpmo.py,sha256=
|
551
|
+
sglang/srt/models/minicpmo.py,sha256=BAiFR78L0j5WDZtOrUf6JgUe9DZs9huIsfQ_4RzUbdA,76319
|
536
552
|
sglang/srt/models/minicpmv.py,sha256=79zZn3co9r7SERatx49EuHRoLWRiy6qeaUFgjDWJo2I,40571
|
537
553
|
sglang/srt/models/mistral.py,sha256=EYifJUUzN2Z2-iL37eJiNZF_DB0H4pa0mKlgYRIxM70,838
|
538
554
|
sglang/srt/models/mixtral.py,sha256=zQHCL_ZMKmLR7jitpEw8n7Rv6xhxUJzSXklsw6auh2E,14965
|
539
555
|
sglang/srt/models/mixtral_quant.py,sha256=-kQw9r8KcLdO8SNN9RKXzrGq9Q2Al9l9cWHi1VrZSRM,15260
|
540
556
|
sglang/srt/models/mllama.py,sha256=jYV5ckyuJN5XU2VXjUgV1i-Yz5rZDQ-6OYsNZvUTJjo,39775
|
541
|
-
sglang/srt/models/mllama4.py,sha256=
|
557
|
+
sglang/srt/models/mllama4.py,sha256=ejuhIVX0MDojdB5EPR-V-Qd_E9st8KYjnjyeArFbdFs,9084
|
542
558
|
sglang/srt/models/olmo.py,sha256=7-q_fA6XXdG7kPUjpUzYkzMUWJobuSjhqjYw9xSUs_c,12671
|
543
559
|
sglang/srt/models/olmo2.py,sha256=azmljhJF4ivcQfUtfsAUxq3ducE4tRKTL6iwe0IKYMg,14327
|
544
560
|
sglang/srt/models/olmoe.py,sha256=TMzt-yB891bvA4X50xL0NjNnFYSx9imlA7N1EG8KNK0,15949
|
545
561
|
sglang/srt/models/phi3_small.py,sha256=UbqZvpwWolXUPd0zbKgbL93yVXUY1n4kXJLgIe_gjaM,15508
|
546
562
|
sglang/srt/models/qwen.py,sha256=xYkVmMZS2uMqWhfndc8EYm0olpKFnggfuMp_6aobVi4,10758
|
547
563
|
sglang/srt/models/qwen2.py,sha256=ab912Yyk0aXOzI-wrxqN-sNF3bTVkNxB8P2uNcOCv9U,16394
|
548
|
-
sglang/srt/models/qwen2_5_vl.py,sha256=
|
564
|
+
sglang/srt/models/qwen2_5_vl.py,sha256=mqgmDHleJN3GmuZG1pUSpIZYKO1omTsa8P5MXYULAGM,22462
|
549
565
|
sglang/srt/models/qwen2_classification.py,sha256=dGrMm4ebd30_lBhHOhaV57ig2iOTx3nqB4GEzsrRIM8,2747
|
550
566
|
sglang/srt/models/qwen2_eagle.py,sha256=Iz0HWL2FgSD3FqoFhfYmbIZeEYkPTJ96lYbkncmHJX4,4644
|
551
567
|
sglang/srt/models/qwen2_moe.py,sha256=bmS2pyHD5zQo5plTCzAo_mjnahVtJ1jaRSURX1PlQC4,18313
|
552
568
|
sglang/srt/models/qwen2_rm.py,sha256=-mQXDEv11p-I1HXgYLTtY6ROem6UYorO958WsDrzsgs,2837
|
553
|
-
sglang/srt/models/qwen2_vl.py,sha256=
|
569
|
+
sglang/srt/models/qwen2_vl.py,sha256=tgES87Rmdl7hqMLAnxYqpWerxK28n5UY7Ma5309TPqs,21408
|
554
570
|
sglang/srt/models/qwen3.py,sha256=reaowGkotYAGHS5zTCWrvnyxtu92QKus19n-2amtMa4,12358
|
555
571
|
sglang/srt/models/qwen3_moe.py,sha256=Tee7oW6Xvo2pV_Q93y-HKykBFiPjo_-YfeIsIelB3hA,15623
|
556
572
|
sglang/srt/models/registry.py,sha256=inKh9iwOp3LFYm3nqujg-OtABClOP-ifc1stA9cZegA,3434
|
573
|
+
sglang/srt/models/roberta.py,sha256=Zgd35och3pW6TYrNeEoeOZ8qPfbFwB3ngThpVWSPBcY,6320
|
557
574
|
sglang/srt/models/stablelm.py,sha256=0x_31uIr3WcWwecdPAI3ek9KkyKBJS7VwknTk2y0gjY,12281
|
558
575
|
sglang/srt/models/torch_native_llama.py,sha256=5tfFSMAXB3ScToqTALtCXa8Oo-qPCJh-KQCNB6QOlNA,19293
|
559
576
|
sglang/srt/models/xverse.py,sha256=DsNVI9JpzN4jj0Ry6aTrj7r-xq5YLOoDX2kH4YLJA-I,14035
|
560
577
|
sglang/srt/models/xverse_moe.py,sha256=7KCM2-j12towDMNvXkuuYiBOmNauH6NG4Ip40x0khqA,16782
|
561
578
|
sglang/srt/models/yivl.py,sha256=oToK7-u5IGO7xwpJIQ7VtudlK6-zPqJX4bt6_wv0SH8,4850
|
562
|
-
sglang/srt/openai_api/adapter.py,sha256=
|
563
|
-
sglang/srt/openai_api/protocol.py,sha256=
|
579
|
+
sglang/srt/openai_api/adapter.py,sha256=7WMplmT0SWJXo5F8s1s3Q9_6WV_cTscMS1Bodbl9Xes,76746
|
580
|
+
sglang/srt/openai_api/protocol.py,sha256=8Iu4t9JlH99QggKl55PYQWTW81u5mpOj0aA-bs44A_c,13621
|
564
581
|
sglang/srt/platforms/interface.py,sha256=hym3iooBB4C8if5hDZezgVN6h4NIOu7sg2ZUBIV6XmM,11246
|
565
582
|
sglang/srt/sampling/custom_logit_processor.py,sha256=tDvoLgLqn-sy1qcY6vSrpbnHCeqbdk0uhMOO-uy4p4E,1099
|
566
583
|
sglang/srt/sampling/sampling_batch_info.py,sha256=4LCowU2bk0TOSfIGpEy90N1SpTsiOKK8Rx1ZYcklUFQ,11988
|
@@ -571,7 +588,7 @@ sglang/srt/sampling/penaltylib/min_new_tokens.py,sha256=rdU_D7RoIcrQPhysNQEzmr4T
|
|
571
588
|
sglang/srt/sampling/penaltylib/orchestrator.py,sha256=XM-Lm6u7gYPtMZrTIc0FR4QxNZxBH5s_Cj82umyCzYk,5721
|
572
589
|
sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=NRh10AJrrQlGJ6S-enGdRefrTrWpyqrSm-aNnyqQNQQ,2119
|
573
590
|
sglang/srt/speculative/build_eagle_tree.py,sha256=lt4sXUehPi26MT2-2Z0VivtF6AP7kirSaEO_u-YJ4J4,11670
|
574
|
-
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=
|
591
|
+
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=NviXdUvowQkV1kLs3eXLlxJx6UZzyQMZH03zCXpsIg4,9291
|
575
592
|
sglang/srt/speculative/eagle_utils.py,sha256=iJYhklXHfDgEKbVB39HkVEea-XTEC60Z_LjIVjkrZQs,28701
|
576
593
|
sglang/srt/speculative/eagle_worker.py,sha256=D4G8hnwtc8xQt1okG4TY9wYSXbKTqGVDAD22AUXW6pA,26824
|
577
594
|
sglang/srt/speculative/spec_info.py,sha256=rhaKG0TzyF9XZYHEWp1jccwTBohSNsUDvxHFtAoOl18,709
|
@@ -579,8 +596,8 @@ sglang/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
|
579
596
|
sglang/test/few_shot_gsm8k.py,sha256=7VLbWl4nCQs1wjtW4q-46jf9jUCycSs5Iw8v7sUSzBw,4284
|
580
597
|
sglang/test/few_shot_gsm8k_engine.py,sha256=QQbrwOX6-cJDD3RZC_e7zPnt6aSo8JdF8X_lRHSjdDM,3886
|
581
598
|
sglang/test/run_eval.py,sha256=9yO0hXZOcn4abEOs96T-XPguDEklK16Ltco0pGF3zCg,4020
|
582
|
-
sglang/test/runners.py,sha256=
|
583
|
-
sglang/test/send_one.py,sha256=
|
599
|
+
sglang/test/runners.py,sha256=vSOl38rVDR3l2ezVCs672vE-LcOA2rJHjlkhLgEjcz8,30260
|
600
|
+
sglang/test/send_one.py,sha256=_l72sRfuXRUldyD3PD63hg_WxNvvhW5unNnbe4XuAwk,4380
|
584
601
|
sglang/test/simple_eval_common.py,sha256=joqrGysuLnJFtzDRIgFkMsRyKUSyjVPFWp0_PHAL3Ik,12378
|
585
602
|
sglang/test/simple_eval_gpqa.py,sha256=8Xt9Bw05c7SZTYrCZgB68OZUqUbLo69ywiyx0bTvSUk,3220
|
586
603
|
sglang/test/simple_eval_humaneval.py,sha256=zmV3xWYc2OrpiT9Dy55RTKZL5DEROD1cJ0NA_-cU5zI,5685
|
@@ -594,13 +611,13 @@ sglang/test/test_custom_ops.py,sha256=2bSo9P5_rJZYFq8Y8IKRimDfFyZZGJluhL7Ngny0Pf
|
|
594
611
|
sglang/test/test_dynamic_grad_mode.py,sha256=L76yUCuk_ymNpXD2CmO8r2GiGjIvD_gtTsuFDs2NolI,1638
|
595
612
|
sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg,3764
|
596
613
|
sglang/test/test_programs.py,sha256=VZ3vXtUDBnXz0M7gFdDH8hXg9Wa0j_qI8CVqjEgRN_E,18877
|
597
|
-
sglang/test/test_utils.py,sha256=
|
614
|
+
sglang/test/test_utils.py,sha256=1U4Jtx_oz_UtS3SSJdqGuh3ujnj2g8pZjN5MYsbsBwI,32164
|
598
615
|
sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
599
616
|
sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
|
600
617
|
sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
|
601
618
|
sglang/test/attention/test_prefix_chunk_info.py,sha256=er0i3KGHMkw-4UZB1GCFd4oYwRcXfU5wpO1ORqpNGGA,7626
|
602
|
-
sglang-0.4.
|
603
|
-
sglang-0.4.
|
604
|
-
sglang-0.4.
|
605
|
-
sglang-0.4.
|
606
|
-
sglang-0.4.
|
619
|
+
sglang-0.4.6.post1.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
|
620
|
+
sglang-0.4.6.post1.dist-info/METADATA,sha256=UTh1TF2jiAdQunwLv7_bmww5_18c4uD7FCaeO-Z3gAs,25361
|
621
|
+
sglang-0.4.6.post1.dist-info/WHEEL,sha256=ck4Vq1_RXyvS4Jt6SI0Vz6fyVs4GWg7AINwpsaGEgPE,91
|
622
|
+
sglang-0.4.6.post1.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
|
623
|
+
sglang-0.4.6.post1.dist-info/RECORD,,
|
File without changes
|
File without changes
|