sglang 0.5.0rc1__py3-none-any.whl → 0.5.0rc2__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 +0 -1
- sglang/srt/configs/model_config.py +1 -0
- sglang/srt/disaggregation/decode.py +0 -1
- sglang/srt/entrypoints/engine.py +2 -2
- sglang/srt/entrypoints/http_server.py +64 -0
- sglang/srt/entrypoints/openai/protocol.py +2 -0
- sglang/srt/entrypoints/openai/serving_chat.py +1 -0
- sglang/srt/entrypoints/openai/serving_completions.py +1 -0
- sglang/srt/layers/attention/flashinfer_backend.py +3 -0
- sglang/srt/layers/attention/flashinfer_mla_backend.py +1 -0
- sglang/srt/layers/attention/triton_backend.py +24 -27
- sglang/srt/layers/attention/trtllm_mha_backend.py +8 -6
- sglang/srt/layers/attention/trtllm_mla_backend.py +10 -3
- sglang/srt/layers/communicator.py +7 -7
- sglang/srt/layers/dp_attention.py +118 -27
- sglang/srt/layers/logits_processor.py +12 -18
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=129,N=352,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=161,N=192,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_0/E=16,N=1024,device_name=NVIDIA_B200.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/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/triton_3_3_1/E=128,N=768,device_name=NVIDIA_H20.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=640,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=128,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_H20-3e,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/multimodal.py +156 -40
- sglang/srt/layers/quantization/__init__.py +5 -32
- sglang/srt/layers/quantization/awq.py +15 -16
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py +0 -1
- sglang/srt/layers/quantization/gptq.py +12 -17
- sglang/srt/layers/quantization/marlin_utils.py +15 -5
- sglang/srt/layers/quantization/modelopt_quant.py +52 -30
- sglang/srt/layers/quantization/mxfp4.py +16 -2
- sglang/srt/layers/quantization/utils.py +52 -2
- sglang/srt/layers/sampler.py +5 -2
- sglang/srt/lora/layers.py +6 -2
- sglang/srt/managers/cache_controller.py +4 -1
- sglang/srt/managers/io_struct.py +14 -0
- sglang/srt/managers/schedule_batch.py +18 -39
- sglang/srt/managers/scheduler.py +3 -4
- sglang/srt/managers/tokenizer_manager.py +28 -18
- sglang/srt/mem_cache/allocator.py +8 -157
- sglang/srt/mem_cache/allocator_ascend.py +158 -0
- sglang/srt/mem_cache/chunk_cache.py +1 -1
- sglang/srt/model_executor/cuda_graph_runner.py +8 -21
- sglang/srt/model_executor/forward_batch_info.py +8 -10
- sglang/srt/model_executor/model_runner.py +57 -53
- sglang/srt/models/deepseek_nextn.py +2 -1
- sglang/srt/models/deepseek_v2.py +5 -3
- sglang/srt/models/glm4_moe.py +2 -2
- sglang/srt/models/glm4_moe_nextn.py +2 -1
- sglang/srt/models/gpt_oss.py +7 -2
- sglang/srt/models/llama.py +10 -2
- sglang/srt/models/llama4.py +18 -5
- sglang/srt/models/qwen2.py +2 -2
- sglang/srt/models/qwen2_moe.py +20 -5
- sglang/srt/models/qwen3_classification.py +78 -0
- sglang/srt/models/qwen3_moe.py +18 -5
- sglang/srt/models/step3_vl.py +6 -2
- sglang/srt/operations.py +17 -2
- sglang/srt/sampling/sampling_batch_info.py +7 -4
- sglang/srt/server_args.py +33 -7
- sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +7 -21
- sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py +7 -21
- sglang/srt/two_batch_overlap.py +4 -8
- sglang/test/test_marlin_moe.py +1 -1
- sglang/test/test_marlin_utils.py +1 -1
- sglang/version.py +1 -1
- {sglang-0.5.0rc1.dist-info → sglang-0.5.0rc2.dist-info}/METADATA +5 -5
- {sglang-0.5.0rc1.dist-info → sglang-0.5.0rc2.dist-info}/RECORD +75 -63
- sglang/srt/layers/quantization/scalar_type.py +0 -352
- {sglang-0.5.0rc1.dist-info → sglang-0.5.0rc2.dist-info}/WHEEL +0 -0
- {sglang-0.5.0rc1.dist-info → sglang-0.5.0rc2.dist-info}/licenses/LICENSE +0 -0
- {sglang-0.5.0rc1.dist-info → sglang-0.5.0rc2.dist-info}/top_level.txt +0 -0
@@ -1,6 +1,6 @@
|
|
1
1
|
sglang/__init__.py,sha256=CK60u1J439CJZaPihWqlDjFZ8doIC-5YgIYz-E1PIh4,1823
|
2
2
|
sglang/bench_offline_throughput.py,sha256=Bge0KG4Uiv9CHsxDnqKdXJmG_TcxM4blz3CmvLmDRvM,14815
|
3
|
-
sglang/bench_one_batch.py,sha256=
|
3
|
+
sglang/bench_one_batch.py,sha256=sFJ9nN-M9PTB1Sy5Xa5b1-L3IzfiEj4rQetSB_YLffg,23398
|
4
4
|
sglang/bench_one_batch_server.py,sha256=LS1BRmFwP67IpYlU5wmkWjqquiA5drvtWe-fwHnFSKI,14170
|
5
5
|
sglang/bench_serving.py,sha256=nOjDnqOKDezDvtCvEhC1_FAoWVkir38bSaEIcSGGVZg,71748
|
6
6
|
sglang/check_env.py,sha256=qDMIG2rCNBH1yKnxQmF-Bp10oiFMUKMgfZLHZYOmdSY,8412
|
@@ -9,7 +9,7 @@ sglang/global_config.py,sha256=ZMTux_PsGnvkyJ0kTFwhTdbnFwIjjpGDogut_9Lu4Vo,1732
|
|
9
9
|
sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
|
10
10
|
sglang/profiler.py,sha256=tEHzHerXC-ymk4OrkoUcMbgcGHmb8VESthsNSP2Yx9w,4417
|
11
11
|
sglang/utils.py,sha256=EF1BDyviT9rj3kVrqZEgax2_gX2gZsWq0HIR7Zf6YtA,16394
|
12
|
-
sglang/version.py,sha256=
|
12
|
+
sglang/version.py,sha256=hVkz5IR8OavTJcQvBQ7pJq8m2jZBCgPJLhJfPncjkj0,25
|
13
13
|
sglang/eval/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
|
14
14
|
sglang/eval/loogle_eval.py,sha256=-CC2s2kh5qUoDrHRkQVkC_jNvBgNojXbf456ny5s78s,4557
|
15
15
|
sglang/lang/api.py,sha256=rcp3GeoyZhmJ0GDLPRkuZNcxd0TBJy_wfUDpcmQoqW8,7210
|
@@ -35,14 +35,14 @@ sglang/srt/custom_op.py,sha256=7OPQAZe3PqDXk-xM0Wxl7pm9cazilNJfkNiYIWztWgI,3198
|
|
35
35
|
sglang/srt/hf_transformers_utils.py,sha256=e--dYjSaaB0t2FhrvHlUrd-l2KmrLIRRtyGErDAGV_Q,13786
|
36
36
|
sglang/srt/jinja_template_utils.py,sha256=XCuqi6iHQrtziUvhy22ePdh8WcS42O9sKxuXpM_bMXc,7836
|
37
37
|
sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
|
38
|
-
sglang/srt/operations.py,sha256=
|
38
|
+
sglang/srt/operations.py,sha256=6Inmt4MQSRrlLxYFYTraVSD9u1cmGY6mk6fSJC9u7-M,5931
|
39
39
|
sglang/srt/operations_strategy.py,sha256=-w6GI2VgSGvORy3pUWRAJo6244ykSW1ZBV1Dx0fZlHA,7138
|
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=MxJ-urjCWfvrBRKhdkionRaiSbZMPmk6QAoY-zVc5qw,22910
|
43
|
-
sglang/srt/server_args.py,sha256=
|
43
|
+
sglang/srt/server_args.py,sha256=BK0WbcFd-4pmTGGI2AHgmeXkKswbKHwDu7EAqvePih4,96728
|
44
44
|
sglang/srt/torch_memory_saver_adapter.py,sha256=K_eTx0UU84MHSTXI3iqYLdHV4IWtJMJ2FKdGFJR8v1E,2417
|
45
|
-
sglang/srt/two_batch_overlap.py,sha256=
|
45
|
+
sglang/srt/two_batch_overlap.py,sha256=D-zWCnJnhkmmqzcePSgGhcn7Wu2PUheyCTP8HdEvwa8,34136
|
46
46
|
sglang/srt/utils.py,sha256=gCpO4VqGbAD_xvp7qAc7XtEBG2n7nuZNsxV5V_IAVkg,93674
|
47
47
|
sglang/srt/warmup.py,sha256=zldxhMlXpclRAJXmfBjJNUJd1eDizVdysibBvQyTVuA,1782
|
48
48
|
sglang/srt/configs/__init__.py,sha256=gas3cKYVgZp7qF4m8FejX4EUUmdQd_w5WDTm7fo7yM0,743
|
@@ -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=
|
59
|
+
sglang/srt/configs/model_config.py,sha256=J8jb35erM0xsvhRVCUQDzZhJIabNu3E9SoIorbFYzW8,30415
|
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
|
@@ -79,7 +79,7 @@ sglang/srt/debug_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3
|
|
79
79
|
sglang/srt/debug_utils/dump_comparator.py,sha256=8ZPwP0zojfRHvPiSKgpESYFcC3tR-sBjIej5SotOmpg,3947
|
80
80
|
sglang/srt/debug_utils/dumper.py,sha256=eyrg5lmwyYF88kn8b06kpu1__aGnWb6CLM8fxywtooI,3158
|
81
81
|
sglang/srt/debug_utils/text_comparator.py,sha256=pg58q3fzsnd3rJKRgZTdcOviOtAv8pSYZdwf-TLffMM,5522
|
82
|
-
sglang/srt/disaggregation/decode.py,sha256=
|
82
|
+
sglang/srt/disaggregation/decode.py,sha256=8uu_IZthGye-g-htPHjr60OtYrQjLnan1hoz6Dsgs-8,34289
|
83
83
|
sglang/srt/disaggregation/decode_schedule_batch_mixin.py,sha256=Iu7ehHGAfYIVATtClwX6y-6zxWs-JyDlMotYo_5yuUc,6598
|
84
84
|
sglang/srt/disaggregation/kv_events.py,sha256=yFbtwOeblcCmOXTjg00TopxnyjkmCBQIVz46KB8jetY,13555
|
85
85
|
sglang/srt/disaggregation/launch_lb.py,sha256=Elw-J-97jTA70Ml1EMp_J-k9X9H5WHTWOC2XEvmyJRY,3827
|
@@ -119,16 +119,16 @@ sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=IrSrnpZnii0E
|
|
119
119
|
sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
|
120
120
|
sglang/srt/entrypoints/EngineBase.py,sha256=yKN76witT2jz1zhmLHmPNLGMpK2UiOTaKQ2KPD8l99U,2594
|
121
121
|
sglang/srt/entrypoints/context.py,sha256=s-kgI8t1t_NbcpWr52JiCqutLx3aiCea-PagiWVQHu0,7373
|
122
|
-
sglang/srt/entrypoints/engine.py,sha256=
|
122
|
+
sglang/srt/entrypoints/engine.py,sha256=y8bf1uh45QIG1x2otFoo9pxG3nlofu6uTDCYDVRQdpc,32119
|
123
123
|
sglang/srt/entrypoints/harmony_utils.py,sha256=01T-A5GBUm2b306PcxNEg2rfx4cykBcqNYrzcXTWBlc,13590
|
124
|
-
sglang/srt/entrypoints/http_server.py,sha256=
|
124
|
+
sglang/srt/entrypoints/http_server.py,sha256=75PA8JjfgsZ0_12VoBBzGDhw1eAfkEErfx3TFb5Tmk8,43951
|
125
125
|
sglang/srt/entrypoints/http_server_engine.py,sha256=_--j4U04OeJLlnnv1f0XmCd_Ry0z1FlhkrbePX8rYV0,4938
|
126
126
|
sglang/srt/entrypoints/tool.py,sha256=6fyXArrLqDsZO3rTtPfZ_6ngYLfTMRyyTSvESqos5tA,2678
|
127
127
|
sglang/srt/entrypoints/openai/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
128
|
-
sglang/srt/entrypoints/openai/protocol.py,sha256=
|
128
|
+
sglang/srt/entrypoints/openai/protocol.py,sha256=agMKqVv4wYKYLZyGBZD26vfJYD0D81iOdJhUhvtrfPw,26499
|
129
129
|
sglang/srt/entrypoints/openai/serving_base.py,sha256=epD6J7K72p6TP8o2dGYf-a76nXeF80vFhPRpWts6684,5445
|
130
|
-
sglang/srt/entrypoints/openai/serving_chat.py,sha256=
|
131
|
-
sglang/srt/entrypoints/openai/serving_completions.py,sha256=
|
130
|
+
sglang/srt/entrypoints/openai/serving_chat.py,sha256=ML-_9l1sDfSl_pxdGPj0J-upvRHpG_uu9S2RWV17CRw,40160
|
131
|
+
sglang/srt/entrypoints/openai/serving_completions.py,sha256=HCzdkXwokcYeq9gsjzTE96Gkvwrw_gV8OpzyqZgj6yM,16770
|
132
132
|
sglang/srt/entrypoints/openai/serving_embedding.py,sha256=hqPt2ELpT1yQ8sBXo801aLJ3sExoPXs_K5ZQtC2vpAs,6274
|
133
133
|
sglang/srt/entrypoints/openai/serving_rerank.py,sha256=8n6y5kC6UhRpWrDtzH0eIp0dznW12FT60MLCR4bvD2U,3333
|
134
134
|
sglang/srt/entrypoints/openai/serving_responses.py,sha256=65KIJ6gjEUz_kbmqejMVY45NixVTz4VqL2zZIMilk3Y,53394
|
@@ -164,19 +164,19 @@ sglang/srt/function_call/step3_detector.py,sha256=SAibrispu0peDQzqOG2-V93KGHtV6U
|
|
164
164
|
sglang/srt/function_call/utils.py,sha256=__ImDF2kNyoLWsYO5RYoryvy1mmgEjnjXlCvLv-uLCM,1695
|
165
165
|
sglang/srt/layers/activation.py,sha256=uhfhh10N5iLoLEBCvFPRjCfDxu5jXe1ehHJb3ISIjig,7840
|
166
166
|
sglang/srt/layers/amx_utils.py,sha256=1mENgHK2B8mgaD1oMtgbZ15Jmy_Uu1QueBmo09Ff2iA,2865
|
167
|
-
sglang/srt/layers/communicator.py,sha256=
|
168
|
-
sglang/srt/layers/dp_attention.py,sha256=
|
167
|
+
sglang/srt/layers/communicator.py,sha256=TLEb4nM4RoGkRLdGlYbz6Fr5-sbG7nHdij594JAG0qM,21133
|
168
|
+
sglang/srt/layers/dp_attention.py,sha256=vnvt5BQxJG4suDmnVRjyXOJKMnb-Q6O_9kKPDZb5AIU,14794
|
169
169
|
sglang/srt/layers/elementwise.py,sha256=MyQUflyKEfPZ-BggW1Kd4hB53RFD6FXGc2S5LXjx_do,16026
|
170
170
|
sglang/srt/layers/flashinfer_comm_fusion.py,sha256=oWnDVaQKOzsI_YoqTJ0ERs-gHGoeF6XqG1j_S0kP4Go,5913
|
171
171
|
sglang/srt/layers/layernorm.py,sha256=Qkx4Jzb3KWOt2okAVM6Is4FDClpUIJShcPGoElUSnEQ,9598
|
172
172
|
sglang/srt/layers/linear.py,sha256=LeAfQqiYSZLv_Dqu7F0UdiiUIQRsJjaW-j-VwvmZ9mk,52462
|
173
|
-
sglang/srt/layers/logits_processor.py,sha256=
|
174
|
-
sglang/srt/layers/multimodal.py,sha256=
|
173
|
+
sglang/srt/layers/logits_processor.py,sha256=B0miOSuVyBsVhzgmkp7LusIhbEaMCjNKgU4TGh4wVZ8,26778
|
174
|
+
sglang/srt/layers/multimodal.py,sha256=vNswe0J0maeBiN8ehlmxfs-YYlrmMllYl6Nc_Y57Neg,5114
|
175
175
|
sglang/srt/layers/parameter.py,sha256=jCg0G-12GZqTa9gGOqEtVCsnnuBGnYg0VmF6pc6oCFg,17455
|
176
176
|
sglang/srt/layers/pooler.py,sha256=uZ6WX1FLMEafZwusyZdm6KuVlIwSjbKrdwk2qzgqNGk,3812
|
177
177
|
sglang/srt/layers/radix_attention.py,sha256=vtzCPfEiWYunglcLfqO9dcATb-LXzU4sFBwIWsWHbiY,3513
|
178
178
|
sglang/srt/layers/rotary_embedding.py,sha256=7D2zy38ZrSU5eIPyD-zygrssSF6uPvhQlQnBS3mHVc4,73069
|
179
|
-
sglang/srt/layers/sampler.py,sha256=
|
179
|
+
sglang/srt/layers/sampler.py,sha256=cgtAtsxcllqhlrxk5ktiBEAeeVLZb-5wWb9NlGzkNYg,11199
|
180
180
|
sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSahyMfA,4063
|
181
181
|
sglang/srt/layers/utils.py,sha256=NEWsf7OlUgcX42WHXxANduDL0ByBm3beirCp8gvHDRs,1437
|
182
182
|
sglang/srt/layers/vocab_parallel_embedding.py,sha256=x0421kNUgnAb5LnkJR6O67vGZ0ikSD-ThluV3QGsZ_8,22673
|
@@ -187,17 +187,17 @@ sglang/srt/layers/attention/cutlass_mla_backend.py,sha256=PB5BjbJHhmTg_t0kXfarTI
|
|
187
187
|
sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
|
188
188
|
sglang/srt/layers/attention/dual_chunk_flashattention_backend.py,sha256=hCATQEhBATZpPhBMF-cCM9r6nxXA0dr3jOT_hTFS_T0,68656
|
189
189
|
sglang/srt/layers/attention/flashattention_backend.py,sha256=LSx908tNxjyFwbWKF4SG6MASltsMHBeLMQYtNeTOCaQ,94565
|
190
|
-
sglang/srt/layers/attention/flashinfer_backend.py,sha256=
|
191
|
-
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=
|
190
|
+
sglang/srt/layers/attention/flashinfer_backend.py,sha256=F9p_DUWsyZAbmy4hXNBNSbf4BxcjrKP9v37EXI6604M,53751
|
191
|
+
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=_upcQy8QmS6NdSEdxmkSOpd778C2KWUl41642K7fy7o,34114
|
192
192
|
sglang/srt/layers/attention/flashmla_backend.py,sha256=5iSic5ho-lkXNas9mR3uLbXbEl-do31gc8gjR7-a79k,20711
|
193
193
|
sglang/srt/layers/attention/hybrid_attn_backend.py,sha256=dwbxscTGMOnwfxxPpmGdbSiQ0kQRHnH0pdpKHA-_vqc,3181
|
194
194
|
sglang/srt/layers/attention/intel_amx_backend.py,sha256=hH9lxXs_o66_IVs58oPhRCOXGUrHbXSz82FKtHr9GO8,4092
|
195
195
|
sglang/srt/layers/attention/merge_state.py,sha256=OnWmNV6LYTUj0y8SCy_IK7A9q6v9bZwR4_XiLOIvbrk,1423
|
196
196
|
sglang/srt/layers/attention/tbo_backend.py,sha256=CBnSDUmDvT-Tbhi9NHNo1LLkgUAcIsWMJgKa-sAIKW8,9318
|
197
197
|
sglang/srt/layers/attention/torch_native_backend.py,sha256=RsI4lYiQJAj-Xy0vKkzbyKOWHAVPOD9b0n0dEehwlos,9496
|
198
|
-
sglang/srt/layers/attention/triton_backend.py,sha256=
|
199
|
-
sglang/srt/layers/attention/trtllm_mha_backend.py,sha256=
|
200
|
-
sglang/srt/layers/attention/trtllm_mla_backend.py,sha256=
|
198
|
+
sglang/srt/layers/attention/triton_backend.py,sha256=BobOz3UfVjzSoFxPSbELDmZGmlN6vR9PgWCAzFPEnvk,40257
|
199
|
+
sglang/srt/layers/attention/trtllm_mha_backend.py,sha256=XsnPenywiHSuneygzziBJq40prmzZ8ECun9EsXB6uBw,12769
|
200
|
+
sglang/srt/layers/attention/trtllm_mla_backend.py,sha256=0i2gWCr1h37z53nQeyVDUgacSzpZJsun92hf8PPUvmk,18568
|
201
201
|
sglang/srt/layers/attention/utils.py,sha256=lx_jNw2VwlBe2VnDb1eXRRdN9WArw9rdWHTHqndUEFk,3174
|
202
202
|
sglang/srt/layers/attention/vision.py,sha256=a9fUE5FH-aUMmT6wWVOSYQALkhswUjJDmAxYkeWnKfo,19473
|
203
203
|
sglang/srt/layers/attention/wave_backend.py,sha256=_rjyALSl5huNtIGSe6OJlBrC04LhwBBu62bgjlkavkw,23010
|
@@ -368,7 +368,9 @@ sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=fC9PGO2Z3XAQ
|
|
368
368
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=768,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=pdQ1RvXvdWDn8Y8-8MAX3vn-T-wbtkZvHV9GZZvNjnc,3266
|
369
369
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=768,device_name=NVIDIA_H200.json",sha256=JraM-Nvbg5V_TJkSl6UPFYZN1zHHoIbr2pAcksenoTY,3248
|
370
370
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=96,device_name=NVIDIA_H20.json",sha256=JtcHRlPz8xQEAqJ9EWI63oYvdmjQFG6VTHqtt85VOSA,3221
|
371
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=129,N=352,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json",sha256=L8bVHdPaB7r5kcNhoEeN-e_GFnz6OcBc8k206EnkB7A,3244
|
371
372
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=160,N=320,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json",sha256=ZRON2XIRRZrvJ3tbPuQ7-0HqGZ_FkPpRN3yj1isGmpg,3255
|
373
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=161,N=192,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json",sha256=eA1hf3skXF7yE3Hy3eDMM3sblzwP2HfTUga-YiPXnJY,3246
|
372
374
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=ttFuL1rKnEpShItmlGyClOfpUwLi5cZYFaVgo7CAxnA,3268
|
373
375
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Pc2ri-qDByslfD8LgJtRB9QDWjI-KqMeA5ftltiwy8k,3262
|
374
376
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=uuua9V6-1H38ASMPQ60t7UOtV0IQLZWBCl13MwQ25Ss,3264
|
@@ -384,12 +386,21 @@ sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=fC9PGO2Z3XAQ
|
|
384
386
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=272,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=Piw4LN6d8QYrUahWsw3XUOtTMD1o3vHPwA94sGI56Gk,3242
|
385
387
|
"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
|
386
388
|
"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
|
389
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_0/E=16,N=1024,device_name=NVIDIA_B200.json",sha256=he873aoOy7KfFg-uMoTFV4IP7Yk0Dk7mOTuLCTqwZZc,3250
|
387
390
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=352,device_name=NVIDIA_RTX_6000_Ada_Generation,dtype=fp8_w8a8.json",sha256=RMYwqtB7hYeaCyuZ1feVnRhBGlN4jMcmNQX2hqUqBlk,3254
|
388
391
|
"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
|
392
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=qc7qnAk3BYcP5m_eT8IPDfnvFrng_kiRCS36_Uy1M2o,3260
|
393
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=768,device_name=NVIDIA_H20.json",sha256=T2sNHVZCLbyuO68ZeudUzQMBsAQzs2-eo53I-EUIuWM,3233
|
389
394
|
"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
|
390
395
|
"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
|
391
396
|
"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
|
397
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=640,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=naKw-2qby2xDupZ1Lmatf2o_CobkdvFS8V6CUORWLpI,3259
|
398
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=OW3W029Obm4HoiuxZbbjTDwTUIGZ4Qr9zDsumYc9m7k,3261
|
399
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=128,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=P0btzx5KorCnqUxYzbubIfUhfavja28d-7pEdiXI0Tg,3267
|
392
400
|
"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
|
401
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=oH4bCkjBe9PhvqfdqTl1Dnnh95HaPT43VDiz7VjkMUA,3260
|
402
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_H20-3e,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=MeULXWyN79rosSTJp3QODmE_yEQHCkcDIvAVN3WOSBY,3258
|
403
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Vz1g-YtTWcv573EYaR7LzhJlRyonaMsHPT2MMjuRJ9k,3270
|
393
404
|
"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
|
394
405
|
"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
|
395
406
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=384,N=256,device_name=NVIDIA_H20-3e,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=941ZXJti99_GUeiylwPUWqYK8b548xuNIjv0pz9Kcqg,3262
|
@@ -402,8 +413,8 @@ sglang/srt/layers/moe/token_dispatcher/__init__.py,sha256=QUjIW1xImUsIf9M8Ze6ger
|
|
402
413
|
sglang/srt/layers/moe/token_dispatcher/base_dispatcher.py,sha256=ibQnHlcnemt8mwOMj01qz360XaherXvTY0T6wGOeBQU,1273
|
403
414
|
sglang/srt/layers/moe/token_dispatcher/deepep.py,sha256=IUkwcp9nQXeyDoqyqrxz2UGnT5pPSpe8nZwaxBZ99VE,23777
|
404
415
|
sglang/srt/layers/moe/token_dispatcher/standard.py,sha256=jseBfZpst5iCDxXxQV_-53qb0VMqy8RHqXJMf86Bdgg,430
|
405
|
-
sglang/srt/layers/quantization/__init__.py,sha256=
|
406
|
-
sglang/srt/layers/quantization/awq.py,sha256=
|
416
|
+
sglang/srt/layers/quantization/__init__.py,sha256=KJ_DGIe-vhd4qTPK8GPNdLDBCdsmoHNa71TlFrfUM_k,7856
|
417
|
+
sglang/srt/layers/quantization/awq.py,sha256=lkSljl4LuBOG8QFsGAFdahGjO9i8KHuWNrSrjkrTqnI,26485
|
407
418
|
sglang/srt/layers/quantization/awq_triton.py,sha256=4XwJbRNlFmlvjbMGtWF1Wvk26Kf96162vNEmuI5Kars,11737
|
408
419
|
sglang/srt/layers/quantization/base_config.py,sha256=CiSxG5iuOa-12V9kIOh0Fu8Y7knZxfFbbNGTo6nEaRE,7173
|
409
420
|
sglang/srt/layers/quantization/blockwise_int8.py,sha256=3avetVUBjCiuYWBJQbElbVXWpx0SgsFsDc5F9F_zHgA,13879
|
@@ -411,27 +422,26 @@ sglang/srt/layers/quantization/fp4.py,sha256=rmitvYd4PqDK906oeRgpaaKs_XIdDLMJ1Uz
|
|
411
422
|
sglang/srt/layers/quantization/fp8.py,sha256=Ee0c8ft9j5uUAhRM84Ng4oRE1hxfkK-YLBN_Fad0fvk,48753
|
412
423
|
sglang/srt/layers/quantization/fp8_kernel.py,sha256=u-8o7zr3zsuD_tszRM_HMaDS8UyQv_Q06kiVa9uyK4U,49779
|
413
424
|
sglang/srt/layers/quantization/fp8_utils.py,sha256=8vvtk-lTfz7-hOm_vbWZk0F6nYqzSaX0k0VHrvkGZj4,26966
|
414
|
-
sglang/srt/layers/quantization/gptq.py,sha256=
|
425
|
+
sglang/srt/layers/quantization/gptq.py,sha256=xgqaqIK2biZ8MI2-I4tYLAPiVif7kOV8HKYEGkBxIlU,39030
|
415
426
|
sglang/srt/layers/quantization/int8_kernel.py,sha256=GfRC9FOn9exNvK4QHbUeBj3Hhv32VcyGphapFPt5b84,12625
|
416
427
|
sglang/srt/layers/quantization/int8_utils.py,sha256=YK9CS-lb_n91kNCTKK5o5apYF31V2giDg5G5VKrpcUA,2356
|
417
428
|
sglang/srt/layers/quantization/kv_cache.py,sha256=_9pF5rwvB7ta6Gdc5YKVVGbNzYwqmhIx4TrX1-xnodQ,3261
|
418
|
-
sglang/srt/layers/quantization/marlin_utils.py,sha256=
|
419
|
-
sglang/srt/layers/quantization/modelopt_quant.py,sha256=
|
429
|
+
sglang/srt/layers/quantization/marlin_utils.py,sha256=m60HSsDaL9X9cI99zjh7JVVeFvlk1i9PIa-Ys0-H80s,25989
|
430
|
+
sglang/srt/layers/quantization/modelopt_quant.py,sha256=fq3OLaq4shaz4ltrAzSpnGdssk54yuGsYg9MRxY7L1k,47358
|
420
431
|
sglang/srt/layers/quantization/moe_wna16.py,sha256=NGcFjiMXwfOX10dkHAY3EgZHxox2k7I402Bvwj3jW1Q,19105
|
421
|
-
sglang/srt/layers/quantization/mxfp4.py,sha256=
|
432
|
+
sglang/srt/layers/quantization/mxfp4.py,sha256=Gf6I81tz6vUDQF5zGvslYLVgKYbiCTZwNvGTtV2q6SQ,25211
|
422
433
|
sglang/srt/layers/quantization/mxfp4_tensor.py,sha256=I1hh_uZqIqkgFnNyjaP17cDMLMV5rs3aFytg91MPhzk,5357
|
423
434
|
sglang/srt/layers/quantization/petit.py,sha256=74Jn0eICq23v_b4rEctDaYTINtbqkdM6IXycT-_FbbI,8954
|
424
435
|
sglang/srt/layers/quantization/petit_utils.py,sha256=-gy4zMhqNoIA1R0n7-5C0efV54jHonCUgPDUUFjGsyM,3245
|
425
436
|
sglang/srt/layers/quantization/qoq.py,sha256=jM96uJfKz3vo1B6SSIrQXNT3vnD3UJrepK2eY-tSQU4,8139
|
426
|
-
sglang/srt/layers/quantization/scalar_type.py,sha256=nBk4THjeRWMGW3hKWRdrjfQ0gwpuBxmYqapDBaRpnr4,12395
|
427
437
|
sglang/srt/layers/quantization/unquant.py,sha256=zLShmKPEMW1OeuiwDzDteE7a2m7YlM7WVw0jWPSBgOQ,14481
|
428
|
-
sglang/srt/layers/quantization/utils.py,sha256=
|
438
|
+
sglang/srt/layers/quantization/utils.py,sha256=ATf7dfF2rMlcO5BHALsVBFsHU1b6cmM0WXEWnNuLH7o,17210
|
429
439
|
sglang/srt/layers/quantization/w4afp8.py,sha256=Il7qrNaIFK9hS7c8Ig7p5pb6g-eZajiHdtnfyBjseC0,11359
|
430
440
|
sglang/srt/layers/quantization/w8a8_fp8.py,sha256=I_W9A4Pnluaf-SaT8zq-tAJTt1acpxfE_kG5mqdc5cE,10250
|
431
441
|
sglang/srt/layers/quantization/w8a8_int8.py,sha256=G2GxKH_EQovudn4DrZVmmpqcdGWe4B7n5tKT_D0JD50,35379
|
432
442
|
sglang/srt/layers/quantization/compressed_tensors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
433
443
|
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=DsPCI82A4rqI6mjKo-WorIKUweppFV7-16Qku-zW9_g,25897
|
434
|
-
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=
|
444
|
+
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=ACOY0Eoh_8aGtnMqqvvew7XP_19k4lLNrmPv8YLta_E,24318
|
435
445
|
sglang/srt/layers/quantization/compressed_tensors/utils.py,sha256=mnUmKWFQUnY8bVoFHUuNVwqsfS-cefeR-ofyaihCXcY,7621
|
436
446
|
sglang/srt/layers/quantization/compressed_tensors/schemes/__init__.py,sha256=HWMTnmrj-mUCRXgcOwnnXLrvrAE-ONdPTSzSImjHCMA,347
|
437
447
|
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_scheme.py,sha256=tdKJC8c3SX8T3z8JL-1YCsg4ftcv55Wxt0vZrYftpX8,1635
|
@@ -598,7 +608,7 @@ sglang/srt/layers/quantization/quark/utils.py,sha256=Aq85sXcnVN2IJnfwCe5t9hfkodc
|
|
598
608
|
sglang/srt/layers/quantization/quark/schemes/__init__.py,sha256=z1LHBPqkxAS6Wp70DWwESOSixWpIhIQwz1IHcWDdURQ,167
|
599
609
|
sglang/srt/layers/quantization/quark/schemes/quark_scheme.py,sha256=b8-Djo-UQAwgxSWViKDoEP5dOSesnEBPyozX-osB3WY,1478
|
600
610
|
sglang/srt/layers/quantization/quark/schemes/quark_w4a4_mxfp4.py,sha256=gMKmrpimBzzzQ1_XlqNKHPhJffqoEdifi8EJ6EdCRhE,3758
|
601
|
-
sglang/srt/lora/layers.py,sha256=
|
611
|
+
sglang/srt/lora/layers.py,sha256=3RwD5TC4zO8A2Haz7ScIT5vc_neU9RHM3DoamKttVkw,10838
|
602
612
|
sglang/srt/lora/lora.py,sha256=lPPEJavrfANXDzdmKEg4hbgmKIiEr3NU7KNJeaaKTB8,7393
|
603
613
|
sglang/srt/lora/lora_config.py,sha256=qDgMTx_69jyJUl29O5FxLzYa0BMhqYVXWXfyyVOvGm0,1684
|
604
614
|
sglang/srt/lora/lora_manager.py,sha256=nM5UTd_HxtZxULRekQ_qzRj5p8AYohlJT78Opf_1XiI,20749
|
@@ -612,16 +622,16 @@ sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=QURCYxHNR8Ls4SQtt3dvdgjvdDVh
|
|
612
622
|
sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=z9BB84Xu_MjwCu5v9Yvji-FkIzfEjDySAFybc4WYv7A,6483
|
613
623
|
sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=Ksova04wgeGsFqGOXWqJtMYaHgyUYcx8VU42BZQOkVA,5129
|
614
624
|
sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=VqCAFvUtq_l-0RGIkx3W_fzD55QcW2FAcqpvSHOBFp4,5115
|
615
|
-
sglang/srt/managers/cache_controller.py,sha256=
|
625
|
+
sglang/srt/managers/cache_controller.py,sha256=spYUdHn05DGexZjvajFcT-cbbYhwvd3UPFuUmZtG_Rw,30508
|
616
626
|
sglang/srt/managers/configure_logging.py,sha256=8sNXZ2z9pBWOwn-X3wyz013Ob8Nbm1zDxRkxoZjH-l4,1633
|
617
627
|
sglang/srt/managers/data_parallel_controller.py,sha256=VvEkidmYCGAFTv6upEHf_tyTvkM8xLhCyOnLJGWL6BQ,14592
|
618
628
|
sglang/srt/managers/detokenizer_manager.py,sha256=ZbCFTgOYT4hTB9zvOobmFI8z7HrdsRmRD7TEq9RqTos,10711
|
619
|
-
sglang/srt/managers/io_struct.py,sha256=
|
629
|
+
sglang/srt/managers/io_struct.py,sha256=Luoy_IYmIA_r7sb1t-hz6BI-qnAswKGPR4ntwzQgQAA,39143
|
620
630
|
sglang/srt/managers/mm_utils.py,sha256=DQDiTbnLp0hYMVdSCrl3T2njMP92orVAI4pso3YheEk,29361
|
621
631
|
sglang/srt/managers/multimodal_processor.py,sha256=cnWpu2G79v1a6FJB_FriLxESgGUbfC3GptLmeRVVgew,1801
|
622
|
-
sglang/srt/managers/schedule_batch.py,sha256=
|
632
|
+
sglang/srt/managers/schedule_batch.py,sha256=d8imQLB8bOWoiPmTJvh41LV3q1ZfqBgEvTSSUeW9Kl4,77279
|
623
633
|
sglang/srt/managers/schedule_policy.py,sha256=vh9BQW9tBv80LW4JApLE6smU2m4gy6cAEI25HVXiS60,22383
|
624
|
-
sglang/srt/managers/scheduler.py,sha256=
|
634
|
+
sglang/srt/managers/scheduler.py,sha256=oiQ_VMLdOPhVqwIGfvk4xNI-xuABwbBEObr35Qu4h3g,107338
|
625
635
|
sglang/srt/managers/scheduler_input_blocker.py,sha256=zP8xU_UmU2H0AB6sEqvivDcDB1QDgTauNDYDIJ7Nez4,3683
|
626
636
|
sglang/srt/managers/scheduler_metrics_mixin.py,sha256=-1IOE30eOoTvvpz-Xg9J4UcamwAJle2-odFJ9vrinLE,9341
|
627
637
|
sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=e_l5TRm05liINSAeZU-YPRpJXcutmEWYV30F1vPaxbs,31345
|
@@ -630,13 +640,14 @@ sglang/srt/managers/scheduler_recv_skipper.py,sha256=8rVQCBjn9f26Th4XbeUS3fEsvo1
|
|
630
640
|
sglang/srt/managers/scheduler_update_weights_mixin.py,sha256=R81fU-0aIJoi5URm9IA-n2sEjvn3phGXOhzJALZjiOM,5467
|
631
641
|
sglang/srt/managers/session_controller.py,sha256=dzlMNZlo20FTSl64QqK7y7pElsdCy8ICOWWBPTBVwgs,6040
|
632
642
|
sglang/srt/managers/template_manager.py,sha256=L3eWAhrQoaFYaEKoj9AA7r212bXcj8RJxb9UbhJf9Jg,11823
|
633
|
-
sglang/srt/managers/tokenizer_manager.py,sha256=
|
643
|
+
sglang/srt/managers/tokenizer_manager.py,sha256=_yUuiY9Pmn0jc3E1by-BGUiFgmKdpgYrhuxhVZHunc4,83080
|
634
644
|
sglang/srt/managers/tp_worker.py,sha256=923H0m12cVt_b5_6uaBhyceGvbGP56mtOX3l5tGnsnM,11904
|
635
645
|
sglang/srt/managers/tp_worker_overlap_thread.py,sha256=NXjrBT7lvdUlDDNFzEQyFYCerpqj92sNl2YynGhQj-Y,11118
|
636
646
|
sglang/srt/managers/utils.py,sha256=RogDdqb2bcxpNk5TbxKyaXFpgFz8t30IOnDp4s-IilM,2776
|
637
|
-
sglang/srt/mem_cache/allocator.py,sha256=
|
647
|
+
sglang/srt/mem_cache/allocator.py,sha256=5sO8TD2CYdNfEsNkjI9bcT9pGEuELPs2Hvj74Y6x-dc,18118
|
648
|
+
sglang/srt/mem_cache/allocator_ascend.py,sha256=1oS_emiKDkigfkSJK2YV2ne-Xj8nUjfHIwJ-gdVpNGg,4734
|
638
649
|
sglang/srt/mem_cache/base_prefix_cache.py,sha256=hLS2ncTMAz7Kpdk5pNwn5c6g8b61_K9OCBL4Vj1xsYc,2801
|
639
|
-
sglang/srt/mem_cache/chunk_cache.py,sha256=
|
650
|
+
sglang/srt/mem_cache/chunk_cache.py,sha256=xQjfmpijkkNRuqfwYMhoiV2nJAviABX4uOSLE-yyVus,3184
|
640
651
|
sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
|
641
652
|
sglang/srt/mem_cache/hicache_storage.py,sha256=Sjl0ZJFj7dqVuinhcoKu3upueVNYWifzGMTPM8exNAE,6375
|
642
653
|
sglang/srt/mem_cache/hiradix_cache.py,sha256=J-k8HYwJuZlKAFxKWSWvDgaAUp-bgANWH8rr46chnCo,28173
|
@@ -660,9 +671,9 @@ sglang/srt/mem_cache/storage/nixl/nixl_utils.py,sha256=i2C70Ct2xdZtR_6OSL7uQk7pe
|
|
660
671
|
sglang/srt/mem_cache/storage/nixl/test_hicache_nixl_storage.py,sha256=TBtN12muLFXZ296mK2d3TSTyj8qhMo5kAYWLpSjLBhY,8017
|
661
672
|
sglang/srt/metrics/collector.py,sha256=FH6Wil2lMK7d7q74CfxfdULCaHKDXeAoo8UBZqqdepM,20289
|
662
673
|
sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
|
663
|
-
sglang/srt/model_executor/cuda_graph_runner.py,sha256=
|
664
|
-
sglang/srt/model_executor/forward_batch_info.py,sha256=
|
665
|
-
sglang/srt/model_executor/model_runner.py,sha256=
|
674
|
+
sglang/srt/model_executor/cuda_graph_runner.py,sha256=JfwUWBu0yXfJaB_uEe1Wit49Hdll6B2ae8JoNT-jKLQ,33416
|
675
|
+
sglang/srt/model_executor/forward_batch_info.py,sha256=3N1SelM99XS2HDWW5DhHrpnRjJH1lYs8hDRtjkCOBaM,40525
|
676
|
+
sglang/srt/model_executor/model_runner.py,sha256=mxqdZCHROajHMQbfEdPBkVKB-cMZZacCMdDzp4wjthk,78323
|
666
677
|
sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
|
667
678
|
sglang/srt/model_loader/loader.py,sha256=pphR7OKRqpsNt7rU0wWg4z-Pakx40a6tvWotfEFY7vU,60131
|
668
679
|
sglang/srt/model_loader/utils.py,sha256=svSrPQxViBz_-0_sByu4coCYrzJ51-4FCi_bFqa6rDo,4474
|
@@ -677,8 +688,8 @@ sglang/srt/models/commandr.py,sha256=5Y_b3K0QY7D37nFGkyiGgY38RleRui_GJUYcHSuHUZo
|
|
677
688
|
sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,15598
|
678
689
|
sglang/srt/models/deepseek.py,sha256=sqobTr9a9c5pNLNJZE-WrJIE_qBq0lC5gCGv9NpGU_Q,17364
|
679
690
|
sglang/srt/models/deepseek_janus_pro.py,sha256=td8xGs6ARfJ8AQCYwUhMOZoWigrAs7m3trF5-kXCqik,70418
|
680
|
-
sglang/srt/models/deepseek_nextn.py,sha256=
|
681
|
-
sglang/srt/models/deepseek_v2.py,sha256=
|
691
|
+
sglang/srt/models/deepseek_nextn.py,sha256=7Z_gO6XkOFM50pQ5oyfuITVVr-a5yKCEb24-csv5W4g,5974
|
692
|
+
sglang/srt/models/deepseek_v2.py,sha256=jMRyz7HBEIlNUJvVmgrpjTZnd0zvXy8h7SvTjUB_1AE,106923
|
682
693
|
sglang/srt/models/deepseek_vl2.py,sha256=7X3gI4DE7guUtNJvMLf06jO8MUHKa2Aan8evZg5hsXc,13061
|
683
694
|
sglang/srt/models/ernie4.py,sha256=_s_Cvw-lwoeSAvGBHSfG1X8_pmPJoJv6e6TvAJGYfoA,16084
|
684
695
|
sglang/srt/models/ernie4_eagle.py,sha256=E7f-ygCcY8pIoyTMKFyVC2uRcG59-_RaQTO4gd5_uWI,7223
|
@@ -692,13 +703,13 @@ sglang/srt/models/gemma3n_audio.py,sha256=isgKfjA5UieYawxU6medL2ssXlzYPqAbagDBnL
|
|
692
703
|
sglang/srt/models/gemma3n_causal.py,sha256=kBDlpSWAg6ZDmUPe08Q7sj5CbdVD_0h07zwJw8sKu7I,36335
|
693
704
|
sglang/srt/models/gemma3n_mm.py,sha256=XTY3HIUt4xv407aS_2rXdfxD3G1sp4z2kVhdC4vnxIQ,20354
|
694
705
|
sglang/srt/models/glm4.py,sha256=dijzVTTgg8zyJ3gga95MKY0WwqwnuAaf-VIV93OFAmc,11433
|
695
|
-
sglang/srt/models/glm4_moe.py,sha256=
|
696
|
-
sglang/srt/models/glm4_moe_nextn.py,sha256=
|
706
|
+
sglang/srt/models/glm4_moe.py,sha256=pCTWGOJtMPtOzotyyXCfYKYV4p45vFzNSQVHtbsmOeY,43150
|
707
|
+
sglang/srt/models/glm4_moe_nextn.py,sha256=EuDWOBSd0oFfdvloePlpzAL81M_L7oz0BiWm74RQioc,5943
|
697
708
|
sglang/srt/models/glm4v.py,sha256=uVdlBBVxQoFOf7-4VVXvhlCBOe2yUYzuYvvUin_5dPM,21279
|
698
709
|
sglang/srt/models/glm4v_moe.py,sha256=dykgo7O99Z4dD_HrhFYuCQJRqBIbKnoM3XXZQR9Ye4Y,17267
|
699
710
|
sglang/srt/models/gpt2.py,sha256=kclhxEs8oJk1KCyhmAqo7rZqecVGGHYkc-a1WZi3aIk,9841
|
700
711
|
sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtYhl4,10307
|
701
|
-
sglang/srt/models/gpt_oss.py,sha256=
|
712
|
+
sglang/srt/models/gpt_oss.py,sha256=YyUprvV-DfgQXe2XYMKmSm67tcBuu9M3C1llO42NwRo,46675
|
702
713
|
sglang/srt/models/granite.py,sha256=8q92shxVPAp_cJDohJATffSGd7Z0Oi-vF5jpY6DlK4s,19840
|
703
714
|
sglang/srt/models/granitemoe.py,sha256=Yr1pEOS_8QopQKBIiDpFpMXZ0Wkc1REyPorM3eDTjNE,13808
|
704
715
|
sglang/srt/models/grok.py,sha256=M23KdbuB3rrJoOG2LkS6EvVts1p0fMfmYAzcbjXQy-Q,28243
|
@@ -710,8 +721,8 @@ sglang/srt/models/interns1.py,sha256=nResXUzfcHZoJie1gHuYwyVir7Ys2fWp3hjzTs6Utvk
|
|
710
721
|
sglang/srt/models/internvl.py,sha256=o6XSAvnMYK5tTPbln9RI9BfRfAQZBE8_xmGPffrznnQ,27838
|
711
722
|
sglang/srt/models/kimi_vl.py,sha256=XulRiufQzOSWZe8-bUl2OEdznvsVU9OBsEYipaX3YkM,12877
|
712
723
|
sglang/srt/models/kimi_vl_moonvit.py,sha256=M5L7AJOJ2Zh2pqLZAq8aRqhbTSlipr8XOoD3ix6o2sU,23908
|
713
|
-
sglang/srt/models/llama.py,sha256=
|
714
|
-
sglang/srt/models/llama4.py,sha256=
|
724
|
+
sglang/srt/models/llama.py,sha256=tdyUon6icwzqo4XTAkFxVA2f82g80AGr9ZezK1uXmkM,28126
|
725
|
+
sglang/srt/models/llama4.py,sha256=CZiLZCJACIz73L9pmDc7gqg6tuAMu2ca_NdKVb-XlKQ,19690
|
715
726
|
sglang/srt/models/llama_classification.py,sha256=4QWTFaUZIFKYZvEzs8bx8VkOZNIwdYCLrnwrdAw4QK0,3108
|
716
727
|
sglang/srt/models/llama_eagle.py,sha256=Ubh_jmtfiOnriwvHgQT0ZGID7JvYvdSi_QGHOIkEgPE,5028
|
717
728
|
sglang/srt/models/llama_eagle3.py,sha256=715soylDE3kOMWskbPy_70BFxaSc5ptaHCS1v3-ViA0,8579
|
@@ -742,21 +753,22 @@ sglang/srt/models/phi4mm_utils.py,sha256=bW3un4x3SHDvSkAOHQEFO286FGdOPeKhBpJCOmN
|
|
742
753
|
sglang/srt/models/phimoe.py,sha256=EGaKFi02MZP20qFrILA4o2SJ0wb1HoJagFTSqVUbJcQ,19223
|
743
754
|
sglang/srt/models/pixtral.py,sha256=ZkZrV0mEfmbBdgRko2AvsMFZbAVskgp_5a5p1xCSRN0,16835
|
744
755
|
sglang/srt/models/qwen.py,sha256=j0OU3JjznWi_TbayI09yYGM-Obov6mQzJO82Hs0qBGw,11868
|
745
|
-
sglang/srt/models/qwen2.py,sha256=
|
756
|
+
sglang/srt/models/qwen2.py,sha256=TuKblSvilyVoq9en9e9YpwzLGkpDeOA6IWJELOYSzZA,22733
|
746
757
|
sglang/srt/models/qwen2_5_vl.py,sha256=LfJj-nwcKyl5XAQcaJrQLyQqwCquGORIK-RVAZrnqwM,22813
|
747
758
|
sglang/srt/models/qwen2_audio.py,sha256=SOOsspU7J6eM-iSkms5op0nWhoxJd6QxmwXU2aQBEEQ,7776
|
748
759
|
sglang/srt/models/qwen2_classification.py,sha256=dGrMm4ebd30_lBhHOhaV57ig2iOTx3nqB4GEzsrRIM8,2747
|
749
760
|
sglang/srt/models/qwen2_eagle.py,sha256=6spFg6JYQGeUXLv2Q5eMkVzZvnqGyvW6SdVBtFMxlTM,4806
|
750
|
-
sglang/srt/models/qwen2_moe.py,sha256=
|
761
|
+
sglang/srt/models/qwen2_moe.py,sha256=sxSYHIkRzJncsrt2K1JfaNGBomEwoJEFm8y8B17UUyk,26466
|
751
762
|
sglang/srt/models/qwen2_rm.py,sha256=-mQXDEv11p-I1HXgYLTtY6ROem6UYorO958WsDrzsgs,2837
|
752
763
|
sglang/srt/models/qwen2_vl.py,sha256=ZRAL32Ymt4ZCBqB0JorXAm4ZBrECZ9EbaXKaTSikkQM,21366
|
753
764
|
sglang/srt/models/qwen3.py,sha256=F74fEfkN-hgeHHwSSHLZQtbsmgHvng-0_WnubyNu0_g,19157
|
754
|
-
sglang/srt/models/
|
765
|
+
sglang/srt/models/qwen3_classification.py,sha256=1Hd4VeO56BMTE7ZU4u4YkOdaqZKdf5_59yOStw3I62s,3045
|
766
|
+
sglang/srt/models/qwen3_moe.py,sha256=8IHuWr-CiH7gp2bfBBW6vb03idfFaMAKvgBr5eYIMjk,33156
|
755
767
|
sglang/srt/models/registry.py,sha256=kKQ5ruD6_c3JxWsykOoxt-hroSlslv-Niz0nYvHrhGg,3779
|
756
768
|
sglang/srt/models/roberta.py,sha256=3k53V2Gbezk3jU-D03thXx8csGn5DxFK01ZQ9WFiWPI,9828
|
757
769
|
sglang/srt/models/siglip.py,sha256=DIVJRwdtpLD2QT1kVPIHw7Bn9BE40xJbkHujDNzkjgA,9892
|
758
770
|
sglang/srt/models/stablelm.py,sha256=0x_31uIr3WcWwecdPAI3ek9KkyKBJS7VwknTk2y0gjY,12281
|
759
|
-
sglang/srt/models/step3_vl.py,sha256=
|
771
|
+
sglang/srt/models/step3_vl.py,sha256=jULEA-i7SHtPr6JaXQBCEqtX4HaKllYxD7IUp3PAL0I,35621
|
760
772
|
sglang/srt/models/torch_native_llama.py,sha256=C1WpLn47q0l7IQa0k-Gvbs914FBcpKzqTIX-uvibq-w,18320
|
761
773
|
sglang/srt/models/transformers.py,sha256=ChqRbUq5YD9oXnXwVP81XRoZmsKc44ouYDNgRsH6zmk,10249
|
762
774
|
sglang/srt/models/vila.py,sha256=3NiRHWOqXnCRjXmswAldsNtiE5zRILbScEsbhUzLZF4,9846
|
@@ -784,7 +796,7 @@ sglang/srt/multimodal/processors/qwen_vl.py,sha256=tZOpz8dabyNvjS4hn0UxLf-EaEZH8
|
|
784
796
|
sglang/srt/multimodal/processors/step3_vl.py,sha256=9OI6UW9ybGbAnYN_S-5jh7TN8BGM_bQkvh1AHBJKzAQ,18350
|
785
797
|
sglang/srt/multimodal/processors/vila.py,sha256=vWgV1lTdjiFiTR8w-7fC66-L_tpOOD59c-c9NljGelE,2192
|
786
798
|
sglang/srt/sampling/custom_logit_processor.py,sha256=AwODYVJdRkcQ8PGtJrhzKsqAgn8XZLQbAmR9fGiQzmc,1608
|
787
|
-
sglang/srt/sampling/sampling_batch_info.py,sha256=
|
799
|
+
sglang/srt/sampling/sampling_batch_info.py,sha256=ntC5ZHNth4uBVel4ZLdDVNC5xf8YS7pPw8k7Banb8rk,14792
|
788
800
|
sglang/srt/sampling/sampling_params.py,sha256=HbVcCeSXgPW19MRB-v0FTG1D-zHDAY6-toxyndSl-zI,6462
|
789
801
|
sglang/srt/sampling/penaltylib/__init__.py,sha256=mtN8grFEcaBUhl4yBHmw8NNirt_i6uKO2cDNLHOpZQE,496
|
790
802
|
sglang/srt/sampling/penaltylib/frequency_penalty.py,sha256=Loc3qjJTksNc5s-DV7QZHjgqoo5pxk7-nZzxwyhD2tQ,2144
|
@@ -792,8 +804,8 @@ sglang/srt/sampling/penaltylib/min_new_tokens.py,sha256=rdU_D7RoIcrQPhysNQEzmr4T
|
|
792
804
|
sglang/srt/sampling/penaltylib/orchestrator.py,sha256=XM-Lm6u7gYPtMZrTIc0FR4QxNZxBH5s_Cj82umyCzYk,5721
|
793
805
|
sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=NRh10AJrrQlGJ6S-enGdRefrTrWpyqrSm-aNnyqQNQQ,2119
|
794
806
|
sglang/srt/speculative/build_eagle_tree.py,sha256=O9LJNaBflgJdWT94D-rGH1gJFJ18nst2oOD8HnA2mZ4,12859
|
795
|
-
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=
|
796
|
-
sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py,sha256=
|
807
|
+
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=9Jf6R-78JLTr8AeF9WkSVY7XR9XWO6rZyNHsEqrRFQE,13899
|
808
|
+
sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py,sha256=QuPI0S85KirTlJArLpJXpywv7XQLborDGLFYwLUBnwk,15459
|
797
809
|
sglang/srt/speculative/eagle_utils.py,sha256=x5MKA2GkecNAXE58KKhPJRy5KxCLhRkD-C_kVSZIyzc,46433
|
798
810
|
sglang/srt/speculative/eagle_worker.py,sha256=9NcNI99SsxbGStXdJLylEsrY5Y5wFc1uLvPpeCgSKBw,39336
|
799
811
|
sglang/srt/speculative/spec_info.py,sha256=rhaKG0TzyF9XZYHEWp1jccwTBohSNsUDvxHFtAoOl18,709
|
@@ -823,8 +835,8 @@ sglang/test/test_deepep_utils.py,sha256=749ysTBGNzh6rYUCJhhZBtZpeD15eWTeNHYCytcv
|
|
823
835
|
sglang/test/test_dynamic_grad_mode.py,sha256=L76yUCuk_ymNpXD2CmO8r2GiGjIvD_gtTsuFDs2NolI,1638
|
824
836
|
sglang/test/test_fp4_moe.py,sha256=0oDgC30EEm7dx8cbnmNpZhsRdDL_2NUT86Q5kDijpmI,10104
|
825
837
|
sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg,3764
|
826
|
-
sglang/test/test_marlin_moe.py,sha256=
|
827
|
-
sglang/test/test_marlin_utils.py,sha256=
|
838
|
+
sglang/test/test_marlin_moe.py,sha256=TOuYlVMx2Lv_IN3AMWsg5aGaxQF472E1B9ev_vvfU2Q,7971
|
839
|
+
sglang/test/test_marlin_utils.py,sha256=kC7IzN364mJT9xw0nKUP1V1OK1hkw6fQTBvD4nRaOz4,5151
|
828
840
|
sglang/test/test_programs.py,sha256=DJIbTChloPwaDiEWHQoSTuKv3BGx0-K8Y5e1mDDTACM,18910
|
829
841
|
sglang/test/test_utils.py,sha256=2vUWNlGaInhG5iLNIUOmULlO65CLzojRpbX2yZamj9w,41155
|
830
842
|
sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
@@ -832,8 +844,8 @@ sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqn
|
|
832
844
|
sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
|
833
845
|
sglang/test/attention/test_prefix_chunk_info.py,sha256=hpoDe2wfSa6RlUbfyri_c0iyBTb35UXGL9I2Xh6jamM,7772
|
834
846
|
sglang/test/attention/test_trtllm_mla_backend.py,sha256=ilrX20-1hVo0nmKN0GTqzKoUmDarTXGS8nWJ6c8RHVY,41792
|
835
|
-
sglang-0.5.
|
836
|
-
sglang-0.5.
|
837
|
-
sglang-0.5.
|
838
|
-
sglang-0.5.
|
839
|
-
sglang-0.5.
|
847
|
+
sglang-0.5.0rc2.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
|
848
|
+
sglang-0.5.0rc2.dist-info/METADATA,sha256=xCApmEOvX65YPgwcJhhPbU7tfRUDVLqfP1mhNhmUECE,27495
|
849
|
+
sglang-0.5.0rc2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
|
850
|
+
sglang-0.5.0rc2.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
|
851
|
+
sglang-0.5.0rc2.dist-info/RECORD,,
|