sglang 0.4.4.post4__py3-none-any.whl → 0.4.5.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 +21 -0
- sglang/bench_serving.py +10 -4
- sglang/lang/chat_template.py +24 -0
- sglang/srt/configs/model_config.py +40 -4
- sglang/srt/constrained/base_grammar_backend.py +26 -5
- sglang/srt/constrained/llguidance_backend.py +1 -0
- sglang/srt/constrained/outlines_backend.py +1 -0
- sglang/srt/constrained/reasoner_grammar_backend.py +101 -0
- sglang/srt/constrained/xgrammar_backend.py +1 -0
- sglang/srt/conversation.py +29 -4
- sglang/srt/disaggregation/base/__init__.py +8 -0
- sglang/srt/disaggregation/base/conn.py +113 -0
- sglang/srt/disaggregation/decode.py +18 -5
- sglang/srt/disaggregation/mini_lb.py +53 -122
- sglang/srt/disaggregation/mooncake/__init__.py +6 -0
- sglang/srt/disaggregation/mooncake/conn.py +615 -0
- sglang/srt/disaggregation/mooncake/transfer_engine.py +108 -0
- sglang/srt/disaggregation/prefill.py +43 -19
- sglang/srt/disaggregation/utils.py +31 -0
- sglang/srt/entrypoints/EngineBase.py +53 -0
- sglang/srt/entrypoints/engine.py +36 -8
- sglang/srt/entrypoints/http_server.py +37 -8
- sglang/srt/entrypoints/http_server_engine.py +142 -0
- sglang/srt/entrypoints/verl_engine.py +37 -10
- sglang/srt/hf_transformers_utils.py +4 -0
- sglang/srt/layers/attention/flashattention_backend.py +609 -202
- sglang/srt/layers/attention/flashinfer_backend.py +13 -7
- sglang/srt/layers/attention/vision.py +1 -1
- sglang/srt/layers/dp_attention.py +2 -4
- sglang/srt/layers/elementwise.py +15 -2
- sglang/srt/layers/linear.py +1 -0
- sglang/srt/layers/moe/ep_moe/token_dispatcher.py +145 -118
- sglang/srt/layers/moe/fused_moe_native.py +5 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=512,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=144,N=512,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1024,device_name=NVIDIA_H200.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=20,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=24,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=264,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/{E=257,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json → E=264,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json } +34 -34
- sglang/srt/layers/moe/fused_moe_triton/configs/E=272,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=272,N=64,device_name=NVIDIA_A800-SXM4-80GB.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=288,N=64,device_name=NVIDIA_A800-SXM4-80GB.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +51 -24
- sglang/srt/layers/moe/fused_moe_triton/layer.py +7 -0
- sglang/srt/layers/moe/router.py +7 -1
- sglang/srt/layers/moe/topk.py +37 -16
- sglang/srt/layers/quantization/__init__.py +13 -5
- sglang/srt/layers/quantization/blockwise_int8.py +2 -0
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py +4 -0
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py +68 -45
- sglang/srt/layers/quantization/fp8.py +28 -14
- sglang/srt/layers/quantization/fp8_kernel.py +130 -4
- sglang/srt/layers/quantization/fp8_utils.py +34 -6
- sglang/srt/layers/quantization/kv_cache.py +43 -52
- sglang/srt/layers/quantization/modelopt_quant.py +271 -4
- sglang/srt/layers/quantization/moe_wna16.py +2 -0
- sglang/srt/layers/quantization/w8a8_fp8.py +154 -4
- sglang/srt/layers/quantization/w8a8_int8.py +3 -0
- sglang/srt/layers/radix_attention.py +14 -0
- sglang/srt/layers/rotary_embedding.py +75 -1
- sglang/srt/managers/io_struct.py +254 -97
- sglang/srt/managers/mm_utils.py +3 -2
- sglang/srt/managers/multimodal_processors/base_processor.py +114 -77
- sglang/srt/managers/multimodal_processors/janus_pro.py +3 -1
- sglang/srt/managers/multimodal_processors/mllama4.py +146 -0
- sglang/srt/managers/schedule_batch.py +62 -21
- sglang/srt/managers/scheduler.py +71 -14
- sglang/srt/managers/tokenizer_manager.py +17 -3
- sglang/srt/managers/tp_worker.py +1 -0
- sglang/srt/mem_cache/memory_pool.py +14 -1
- sglang/srt/metrics/collector.py +9 -0
- sglang/srt/model_executor/cuda_graph_runner.py +7 -4
- sglang/srt/model_executor/forward_batch_info.py +234 -15
- sglang/srt/model_executor/model_runner.py +49 -9
- sglang/srt/model_loader/loader.py +31 -4
- sglang/srt/model_loader/weight_utils.py +4 -2
- sglang/srt/models/baichuan.py +2 -0
- sglang/srt/models/chatglm.py +1 -0
- sglang/srt/models/commandr.py +1 -0
- sglang/srt/models/dbrx.py +1 -0
- sglang/srt/models/deepseek.py +1 -0
- sglang/srt/models/deepseek_v2.py +248 -61
- sglang/srt/models/exaone.py +1 -0
- sglang/srt/models/gemma.py +1 -0
- sglang/srt/models/gemma2.py +1 -0
- sglang/srt/models/gemma3_causal.py +1 -0
- sglang/srt/models/gpt2.py +1 -0
- sglang/srt/models/gpt_bigcode.py +1 -0
- sglang/srt/models/granite.py +1 -0
- sglang/srt/models/grok.py +1 -0
- sglang/srt/models/internlm2.py +1 -0
- sglang/srt/models/llama.py +13 -4
- sglang/srt/models/llama4.py +487 -0
- sglang/srt/models/minicpm.py +1 -0
- sglang/srt/models/minicpm3.py +2 -0
- sglang/srt/models/mixtral.py +1 -0
- sglang/srt/models/mixtral_quant.py +1 -0
- sglang/srt/models/mllama.py +51 -8
- sglang/srt/models/mllama4.py +227 -0
- sglang/srt/models/olmo.py +1 -0
- sglang/srt/models/olmo2.py +1 -0
- sglang/srt/models/olmoe.py +1 -0
- sglang/srt/models/phi3_small.py +1 -0
- sglang/srt/models/qwen.py +1 -0
- sglang/srt/models/qwen2.py +1 -0
- sglang/srt/models/qwen2_5_vl.py +35 -70
- sglang/srt/models/qwen2_moe.py +1 -0
- sglang/srt/models/qwen2_vl.py +27 -25
- sglang/srt/models/stablelm.py +1 -0
- sglang/srt/models/xverse.py +1 -0
- sglang/srt/models/xverse_moe.py +1 -0
- sglang/srt/openai_api/adapter.py +4 -1
- sglang/srt/patch_torch.py +11 -0
- sglang/srt/server_args.py +34 -0
- sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +4 -4
- sglang/srt/speculative/eagle_utils.py +1 -11
- sglang/srt/speculative/eagle_worker.py +6 -2
- sglang/srt/utils.py +120 -9
- sglang/test/attention/test_flashattn_backend.py +259 -221
- sglang/test/attention/test_flashattn_mla_backend.py +285 -0
- sglang/test/attention/test_prefix_chunk_info.py +224 -0
- sglang/test/test_block_fp8.py +57 -0
- sglang/test/test_utils.py +19 -8
- sglang/version.py +1 -1
- {sglang-0.4.4.post4.dist-info → sglang-0.4.5.post1.dist-info}/METADATA +14 -4
- {sglang-0.4.4.post4.dist-info → sglang-0.4.5.post1.dist-info}/RECORD +133 -109
- sglang/srt/disaggregation/conn.py +0 -81
- {sglang-0.4.4.post4.dist-info → sglang-0.4.5.post1.dist-info}/WHEEL +0 -0
- {sglang-0.4.4.post4.dist-info → sglang-0.4.5.post1.dist-info}/licenses/LICENSE +0 -0
- {sglang-0.4.4.post4.dist-info → sglang-0.4.5.post1.dist-info}/top_level.txt +0 -0
@@ -1,17 +1,17 @@
|
|
1
1
|
sglang/__init__.py,sha256=8J5PHcfRJul4R1NJnt0BtW05aVQNWqp8PyalLPOTZCA,1669
|
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=upXSMMlhO53J2cAP9lAb3KWt0nTEUdkNRWGk0lobS08,18887
|
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=ek6D6uw0IlsMb0lhg57rBq7q4au7Os78GUMXopy0Wfk,57702
|
7
7
|
sglang/check_env.py,sha256=76itNLUw9KlqbiY1BI4u4YaMZaqyCNcrCLUIb6aHflM,8396
|
8
8
|
sglang/global_config.py,sha256=xzLdk8W53fneFblNh8iIjGF9C3-7mnzR1-LleD9Btxg,1495
|
9
9
|
sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
|
10
10
|
sglang/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
|
11
11
|
sglang/utils.py,sha256=GIcgiRHkZ-gyPxXOdn1qFF41jkg4-YdDxbPc4mzO-qk,16159
|
12
|
-
sglang/version.py,sha256=
|
12
|
+
sglang/version.py,sha256=mfIdQ0Yo6GF2VyWIDGyBUw_42D590eNsz05qnm3UXM4,28
|
13
13
|
sglang/lang/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
14
|
-
sglang/lang/chat_template.py,sha256=
|
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=o1C6G3TzhjSlsH-doTPy5oiVehr57dxNTa5oZw5TTAI,7639
|
17
17
|
sglang/lang/interpreter.py,sha256=OH1SFCm4rUCPO32MTo8j5V2Z13Jic7_r1GQOP1-aHaw,33234
|
@@ -27,18 +27,18 @@ sglang/lang/backend/vertexai.py,sha256=O-iBLD-y3vq80UxnrAoJri7bxpgd-_eakZ88Cf8bE
|
|
27
27
|
sglang/srt/_custom_ops.py,sha256=lUBwC5R2UfjFMA1EtC5Kh2IngsqBJM9IuMW46kJWcjE,3647
|
28
28
|
sglang/srt/aio_rwlock.py,sha256=6LYtOdeTUY3hkfa1dmYkgsaF2ttrwIF3hUWz2AZ2fqw,2970
|
29
29
|
sglang/srt/code_completion_parser.py,sha256=HhEUzdL-FVBsOot9tKDKA1l8Gdx8qsF1RRg-zHNpmLQ,5400
|
30
|
-
sglang/srt/conversation.py,sha256=
|
30
|
+
sglang/srt/conversation.py,sha256=WP72AZrZpiqc5RowucT2tW3jVCb1pb4veW_kpwYS4yY,28785
|
31
31
|
sglang/srt/custom_op.py,sha256=bIZ__3FiZvkbsN9O_jeLy_49X7ZbYbw0VxoL80uWwaI,3715
|
32
32
|
sglang/srt/function_call_parser.py,sha256=buYENeNEP5bhsvD424yGCa9wOqSfVOZSRn6zLiSJp5I,23733
|
33
|
-
sglang/srt/hf_transformers_utils.py,sha256=
|
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
|
-
sglang/srt/patch_torch.py,sha256=
|
36
|
+
sglang/srt/patch_torch.py,sha256=OUPCGQSQz3MVZB1zZ_Eq8lXiw0uIKJ_HWjqQolI8FsM,3088
|
37
37
|
sglang/srt/reasoning_parser.py,sha256=45xsU9RCPfyG4_Zx4y3-JPyNgAtrqwKI4j5R2NT4g1s,5594
|
38
38
|
sglang/srt/server.py,sha256=PrQb9r6L9syWHKlggbbiQYsKtpwSmECqozRbf8qnoV8,874
|
39
|
-
sglang/srt/server_args.py,sha256=
|
39
|
+
sglang/srt/server_args.py,sha256=k42YCDTbEEZZShmoaorQGNRwMxYACDSpvGW2toTb2DQ,53778
|
40
40
|
sglang/srt/torch_memory_saver_adapter.py,sha256=KG3wM9-xZsSdsmORofArnNR7hH55GEyFxaderCDcK9w,1853
|
41
|
-
sglang/srt/utils.py,sha256=
|
41
|
+
sglang/srt/utils.py,sha256=KAEwcTWJZInclD6tGvIEDcXpVW726l6n86FcbrHoCng,60992
|
42
42
|
sglang/srt/warmup.py,sha256=FmJiYfjRr3X_eAe7ojQaPoN17LvHpjDmRWRnO-k86AQ,1469
|
43
43
|
sglang/srt/configs/__init__.py,sha256=vulncVn70WqIT6s0HaB8p_Q6FjOiaLwNZWpoJS9FIuQ,399
|
44
44
|
sglang/srt/configs/chatglm.py,sha256=j-b0YkdYUmQm2y1kNmMJtKeACxWKmBbvNNkDWbs6kbI,2907
|
@@ -48,7 +48,7 @@ sglang/srt/configs/device_config.py,sha256=kfmpPOECqYxcRoY-ko0QZRhyiBWUGP2CMF51D
|
|
48
48
|
sglang/srt/configs/exaone.py,sha256=Duxd4yQoKy8GWEzZD_kCY_OzmN_67CTJL_Kgn0eXk3g,10731
|
49
49
|
sglang/srt/configs/janus_pro.py,sha256=-QtJ4ZGZiAJb0AkOEcuCHzIKLw23nF8nRk3rdCcoUO0,19016
|
50
50
|
sglang/srt/configs/load_config.py,sha256=qs-AxuplouBx2tsv9KGBOLZPbwzuVA4vbktbGP_cRp8,3309
|
51
|
-
sglang/srt/configs/model_config.py,sha256
|
51
|
+
sglang/srt/configs/model_config.py,sha256=oMXM1CardGkJThm0KgCxxgYnwXaJXFZzmUZIM_wugeU,22046
|
52
52
|
sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
|
53
53
|
sglang/srt/connector/__init__.py,sha256=czLX5JOxuMhH-T9eSJzoc1qv1B4z9chyffDRL5I6wo4,1247
|
54
54
|
sglang/srt/connector/base_connector.py,sha256=i6i1TIzsz4NbSEkrdMPq-urb2sN2aLAx8dazga4gB9U,2833
|
@@ -58,16 +58,21 @@ sglang/srt/connector/utils.py,sha256=isTvxauz1_8no5MW6p8Bwj2H9mQkweaRO_RSwAPA7R8
|
|
58
58
|
sglang/srt/connector/serde/__init__.py,sha256=MvBJ7FBJtnou-AD-EdsCdAIDCcA8kWvUNuRViILVJ-0,718
|
59
59
|
sglang/srt/connector/serde/safe_serde.py,sha256=Iv_mFsDvk-cXuw0WTykBZ2yGxW0jb82DwVqC7r3RmAU,750
|
60
60
|
sglang/srt/connector/serde/serde.py,sha256=n59I2MXLa7WCyN_8pEd8L-scJk7lMhmEX-GOUIhF0ZA,1004
|
61
|
-
sglang/srt/constrained/base_grammar_backend.py,sha256=
|
62
|
-
sglang/srt/constrained/llguidance_backend.py,sha256=
|
63
|
-
sglang/srt/constrained/outlines_backend.py,sha256=
|
61
|
+
sglang/srt/constrained/base_grammar_backend.py,sha256=ljTVWpBo3bolce-E_-mtHIY2XWez4qcyDPeaIeZyIhM,7454
|
62
|
+
sglang/srt/constrained/llguidance_backend.py,sha256=Kgd-PQVBQlKWsz506OpF_xSdNBhEbvFywzICTZg21iM,5729
|
63
|
+
sglang/srt/constrained/outlines_backend.py,sha256=XbmkZSJzJnnY7k11uj8Et3StfuOiFwRs3ID4IRYAA4Q,6839
|
64
64
|
sglang/srt/constrained/outlines_jump_forward.py,sha256=iZWXeR3gNYoMubLGyFmLPO4V2YsN5DiGjD71Xk9iFaE,6418
|
65
|
-
sglang/srt/constrained/
|
66
|
-
sglang/srt/
|
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=
|
65
|
+
sglang/srt/constrained/reasoner_grammar_backend.py,sha256=XFxdZqvPofmtCeIMqR10NOyph06HwbdXfiVI8rIoV5s,3646
|
66
|
+
sglang/srt/constrained/xgrammar_backend.py,sha256=Xf4CiU30XCa_RM4bgFkCw1yLeH4wijfewUHIYjnkv-k,6247
|
67
|
+
sglang/srt/disaggregation/decode.py,sha256=qV0TU_nasIWTCXAngB7g3t5uJ1a8nmYW-KGXdIzaG0I,18624
|
68
|
+
sglang/srt/disaggregation/mini_lb.py,sha256=ZU4M7ZtdKUDzpmNMT_NDFnTdbHzoGIxXSHW2PLdnoLs,7511
|
69
|
+
sglang/srt/disaggregation/prefill.py,sha256=AeO4VcCKJ6X3c-GVY81G2aGA0bz9nNXiYNRXFOBWzWo,9954
|
70
|
+
sglang/srt/disaggregation/utils.py,sha256=gbJIFpYM8XpW4aTThPGhny79jl9aBxOIiT2swJpS_Y8,2017
|
71
|
+
sglang/srt/disaggregation/base/__init__.py,sha256=KR8xXoRCDAy2U623mfP6ujXu42m1_F9EiudjrKu2I_A,130
|
72
|
+
sglang/srt/disaggregation/base/conn.py,sha256=gpf32bhYXWm_iaYB6WcrDaJ-UoL1ZzPI_xpi5pMhRQo,2443
|
73
|
+
sglang/srt/disaggregation/mooncake/__init__.py,sha256=1vacEHmWjf7zgbMPzsXKB08FqNKNCquJdUiDlO41BOk,122
|
74
|
+
sglang/srt/disaggregation/mooncake/conn.py,sha256=LkM9X7Rf3H4hjfSUQh4Yq7icvryTrPg_tszb3_hakcg,22210
|
75
|
+
sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=qy-0HYLhaz90ompOtcOvB5jZhI97iAH1tl7dNvbleGc,3457
|
71
76
|
sglang/srt/distributed/__init__.py,sha256=jFOcyt-wFAPMBUAf9zkZalNQlt-4rqmT6pCKBz1E4qo,149
|
72
77
|
sglang/srt/distributed/communication_op.py,sha256=IBnFUdMftK_VSTMMMitGveonorFUUVNL4guqO31cMSc,1130
|
73
78
|
sglang/srt/distributed/parallel_state.py,sha256=hoTgLYfHIKMb_tSwBTauuusJZ8oY9BsiubTTOF8UfIw,50713
|
@@ -80,47 +85,49 @@ sglang/srt/distributed/device_communicators/pynccl.py,sha256=G-Dut_QJHOUG0j7--Zq
|
|
80
85
|
sglang/srt/distributed/device_communicators/pynccl_wrapper.py,sha256=LblisImY9d6EMz-oPS9J16WHo2Q_SRL1DtlJKK63Hfg,15349
|
81
86
|
sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=bbruDIM1GgKIdB6gi71_I0mpB179I-qyvwKuSj1Kaic,20816
|
82
87
|
sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
|
83
|
-
sglang/srt/entrypoints/
|
84
|
-
sglang/srt/entrypoints/
|
85
|
-
sglang/srt/entrypoints/
|
88
|
+
sglang/srt/entrypoints/EngineBase.py,sha256=xoyvp6XAeDLY2_Q2Ng33H-fRhrXHv2ldJJKd-HuDhqE,1870
|
89
|
+
sglang/srt/entrypoints/engine.py,sha256=cO-Yq5i_hrn_yaAuhkHKkUUVXQmHXcTV4B-l76LjbwU,22137
|
90
|
+
sglang/srt/entrypoints/http_server.py,sha256=wYjyyiajP6SWa3auZHZIUJv30zioB0IwdFKXHlyT5zo,28431
|
91
|
+
sglang/srt/entrypoints/http_server_engine.py,sha256=ihA6y3GXRs28Y9U3SgdQcJQjnw_SVIby7QrVgiafX04,4846
|
92
|
+
sglang/srt/entrypoints/verl_engine.py,sha256=sqQKt-HnEdCfR7CkiyskY7E_BeQMxASMDe_Hq91ni1I,6949
|
86
93
|
sglang/srt/layers/activation.py,sha256=1ykXZO0BGz7DFVE-EK26b02I5AgH2IuU4PQB6oUcF4M,6003
|
87
|
-
sglang/srt/layers/dp_attention.py,sha256=
|
88
|
-
sglang/srt/layers/elementwise.py,sha256=
|
94
|
+
sglang/srt/layers/dp_attention.py,sha256=Tfw2BydGPAeLcDlyl0jzBLOtkiygJhgAcf63RGZhrEE,7535
|
95
|
+
sglang/srt/layers/elementwise.py,sha256=XCrR2i-9dP-H6jQo2zUuquwZrsl_wEQqj5Wxk6WUf7o,13987
|
89
96
|
sglang/srt/layers/layernorm.py,sha256=189bORMggKhYcEYEjl6JRcuIoUPllHo3SheoH6YiORY,4546
|
90
|
-
sglang/srt/layers/linear.py,sha256=
|
97
|
+
sglang/srt/layers/linear.py,sha256=etyzpgJbOUWCiu8abUovZQ8IFJ3ObeqOVHocBzFZOP0,51517
|
91
98
|
sglang/srt/layers/logits_processor.py,sha256=Vp8ibljVEezTr54xzeOcjiJR7JdYO8ItkO5nLIIMVu0,24206
|
92
99
|
sglang/srt/layers/parameter.py,sha256=0OTMtmsNds42e3z3wHTRJiUfxCWFwSL6DHrqgeTgGt8,15151
|
93
100
|
sglang/srt/layers/pooler.py,sha256=rj2lygvleBnyLCBZ8I11HGMgpfIDsT0l3PIkshJwdu4,1606
|
94
|
-
sglang/srt/layers/radix_attention.py,sha256=
|
95
|
-
sglang/srt/layers/rotary_embedding.py,sha256=
|
101
|
+
sglang/srt/layers/radix_attention.py,sha256=VeE8wX-8eNaSfw2JYzrek7EjgdVltEujnkMa_u92hCc,2865
|
102
|
+
sglang/srt/layers/rotary_embedding.py,sha256=qvRNHWJxsyz897ntsEEr9iTj9ZNV_HgWUH7PvBE9ydw,45738
|
96
103
|
sglang/srt/layers/sampler.py,sha256=yipSyN5UWGwGS-BC-WzWMmelys4CCDtK_8b1OpaK6sM,11622
|
97
104
|
sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSahyMfA,4063
|
98
105
|
sglang/srt/layers/vocab_parallel_embedding.py,sha256=QUxd4sELx6p3dHvEKmccPZ-phdd_9EjNdwjH3SJ9zxI,22238
|
99
106
|
sglang/srt/layers/attention/base_attn_backend.py,sha256=X_GIbQuU9njtUEGdUP7E_KRhmGxj3UyPHNESlL3QaQ8,3264
|
100
107
|
sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
|
101
|
-
sglang/srt/layers/attention/flashattention_backend.py,sha256=
|
102
|
-
sglang/srt/layers/attention/flashinfer_backend.py,sha256=
|
108
|
+
sglang/srt/layers/attention/flashattention_backend.py,sha256=QKr_X_7fp96cbghJJj3qlzfN_ZeIZ-0a4wUMcSwwj-o,49327
|
109
|
+
sglang/srt/layers/attention/flashinfer_backend.py,sha256=QqHbVoXp9LqoVvoGebXoo9GcrU7LaRRRh5sG93Daa0s,46277
|
103
110
|
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=pnVhvVEK87iFW8gUb1G7X7c1tqro8R2DSEOFCnlV8Bo,30301
|
104
111
|
sglang/srt/layers/attention/flashmla_backend.py,sha256=1RPFNtQOBw6BWxIjrzfJgA9Nx92udLbR-S5KXmqjxS8,10536
|
105
112
|
sglang/srt/layers/attention/torch_native_backend.py,sha256=KABmBrMqKa4x08kkQYdIcZUGydvmaVJIUfo3y8jhFHI,9270
|
106
113
|
sglang/srt/layers/attention/triton_backend.py,sha256=cyxOaUU1CNhaEezJH9j0dd20cwxwIVGGN3jNXFTVkIY,26714
|
107
114
|
sglang/srt/layers/attention/utils.py,sha256=J9mA-cbZT3uTlaKXo0HEAaeMei_TS2o4McTna9LVDCE,2750
|
108
|
-
sglang/srt/layers/attention/vision.py,sha256=
|
115
|
+
sglang/srt/layers/attention/vision.py,sha256=H7dQofAlZjC48Dr6bo4HBsZSLDhdrbHKChyGejeaEGo,11886
|
109
116
|
sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=DPu_aCPgwPqKWZPEQmp_xA7MPbpV2ip-MEICCB470Ao,19120
|
110
117
|
sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=BXUY8ARHBF2s9x9waiEwfZwcMgvuaJA0gxb4OeUZ_tY,31167
|
111
118
|
sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=hbBvnhc2zqu-E3HNROVXyNOZbtDkVRuFus-yTjmE0Sg,13668
|
112
119
|
sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=Y66gZ37u0GKMPtI8n5MbO6uOxRuGEmKIG0IPbJTOqAM,6213
|
113
120
|
sglang/srt/layers/attention/triton_ops/rocm_mla_decode_rope.py,sha256=664WnAJ91EiCUZOcnVDfbTQf4uGJ4ZDZB1CbxpEUFZc,13866
|
114
|
-
sglang/srt/layers/moe/fused_moe_native.py,sha256=
|
115
|
-
sglang/srt/layers/moe/router.py,sha256=
|
116
|
-
sglang/srt/layers/moe/topk.py,sha256=
|
121
|
+
sglang/srt/layers/moe/fused_moe_native.py,sha256=bf0po921lY9xnlZivdJly0bGIYFlLqp5v8Mz7tG5bdg,4451
|
122
|
+
sglang/srt/layers/moe/router.py,sha256=5Aeqoix_AS4uymb665OJE904wVSBkQeFdZP4e7KKPvg,10530
|
123
|
+
sglang/srt/layers/moe/topk.py,sha256=xjkfC50rapfX27nI2078SoKGXAo82waQVArPVDH8Ehc,10843
|
117
124
|
sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
118
125
|
sglang/srt/layers/moe/ep_moe/kernels.py,sha256=ijqRzS-tb0LGnDU5hW-g0JH104ppADrWaUIDGxb9Feo,22919
|
119
126
|
sglang/srt/layers/moe/ep_moe/layer.py,sha256=1TmWnxv-bW1Qbgru-V-vGnt3ruuTIwHQy0Y5ZA_xzvE,36824
|
120
|
-
sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=
|
127
|
+
sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=zQV7Qr-Zrcr3D3efVvZepRQM02bj5djHPsijPssavk8,20430
|
121
128
|
sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=h9yMFAL_bagUf-qBED8gSWdCOb7d8IdA-pE-L_nIg8E,842
|
122
|
-
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=
|
123
|
-
sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=
|
129
|
+
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=6WbcChGdZmSXl5_WlpC0w1cn_QH69mvICQt0pMBO_nk,55474
|
130
|
+
sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=43-UL9KEMoaiC0cRSzWFbg2PADtcoxfZqjZ6TOvQ7Vk,24551
|
124
131
|
"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
|
125
132
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=JJN0hryyLr5Zv3dSS7C8cPFhAwTT6XxUVnBGMZvV6JA,2752
|
126
133
|
"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
|
@@ -132,6 +139,10 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=CYT3ujh5ifonhqQc1uYSa6maJ
|
|
132
139
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=yf33YmWlVSjjyg0Q4OMAWvc9gjRxvttMrQBUEOfPl4I,4153
|
133
140
|
"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
|
134
141
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=C65Q2Mv1LxFQ_qDnv11IZ9nwl7sGZo72nWDflMttu4g,4147
|
142
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=512,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=FsWbV4Q6AzAtgegVuENBDz2ZcSJsqNiwUIVfQbpP7hQ,3244
|
143
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=144,N=512,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=pk6VZChh2Y0CsJSzjtUhOnlta1QLTUEWy33aKQU47XY,3244
|
144
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=Gmk24hc5lVIfQtqSa5wLOcWKedMN8aZUe93DBh6J1AY,3249
|
145
|
+
"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
|
135
146
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1344,device_name=NVIDIA_A100-SXM4-40GB.json",sha256=pCCKkdUzzuBVtljyk7AEIAbeDf12DUiieXaODZXzm5E,3254
|
136
147
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1344,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=trX2-c4N6hTTD6zFNi6A2bT3FkhxKjkM2rPl-o1K9ss,3250
|
137
148
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1344,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=I4d56uD7E1JMXD9RAxq3FebdPquDsnNEkVaIY9Ctm9w,3246
|
@@ -139,6 +150,7 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=CYT3ujh5ifonhqQc1uYSa6maJ
|
|
139
150
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=0ejgnIQ_mzJhKjQpnT-I1Vj9-rPfGlTcQ8u0cXgekUw,2746
|
140
151
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=lU4fjngJmpzOafi-3_q0vj2pLfZQVVagFnZNoI97etk,4128
|
141
152
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=ilUf76TmbtaL3W0bara6JIBEiV_iPIs29UjLbH_GYtc,4145
|
153
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=4ye1VakIIzrGUTxGNb_DM5v59djLgok8SeBeU_YI6Go,3252
|
142
154
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=2688,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=G4PKqWxh0MlBhg7QHKj0m--_fP3Ll0gs7VJaeg-NIDM,3254
|
143
155
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=2688,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=bKX9AvcxN6k-i3RUmHSchZZ3rjoYRYb4iBqhCI4L3MY,3257
|
144
156
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=3072,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=OeJUCPVBD1z9CSooZpy3hRyAasjKqFAQaTLcWK6PWno,2741
|
@@ -152,12 +164,15 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=CYT3ujh5ifonhqQc1uYSa6maJ
|
|
152
164
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json",sha256=kklgf2qLI5CQYiJJ5e9Gxx2gAfGxcyMDYpdJnIXPV8E,2748
|
153
165
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=800,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json",sha256=8e0tN_DHPwvh_HECVHx9oOF_4WWdaht4s6Nmd_K-aBU,2904
|
154
166
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=160,N=192,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=fRzKfVFIcnxqu6DvGJQNltuFRRGz8F-eaL73bIzBzo8,3255
|
167
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=20,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=zq1x3KwZ3YuzE42OPPsElXT_VS9sK135CdB1RsotLU4,3252
|
168
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=24,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=lJbEa7VsBVKL_iRoHniB-2nceMrPctkqtjKszPnX4pk,3248
|
155
169
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json",sha256=RuUDK9XfgXs1eZESWQR9ba4tu-rCRG_UCYwjaJ568sI,3264
|
156
170
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8.json",sha256=wlCi9aoYp7Zc1GThEutvWDbse0kKnNaQgFJsd_L8be0,3259
|
157
171
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json",sha256=tPYxeo_xUOkjQrZMdf9v4IaFrw0RGaZNLGLJPOhjE_g,3260
|
158
172
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json",sha256=MW7KzNa7DcKm53u2Jh-mnb93A3ICefgQHkdKONJMfew,3255
|
159
173
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=0aSYzpv_cBAlpWCPrfGgNTCfae1KdKQnT56E8XFQl7A,3262
|
160
174
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_H20,block_shape=[128, 128].json",sha256=JEOXj48phwoumZWBwNq9TpqxVwIrbnfot4QfAdzvLJI,3249
|
175
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Lqom_VMIPduSZTZQdeL2Wl_x3r9q6RmI9bojJrYwQZ4,3255
|
161
176
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=XmKFaMheq7NNrsvYCJteul0w809l_l460ZiDQC9ToGs,3262
|
162
177
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=256,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=5ro9O8Nf-7MB8NGCQ3QV5kB2k3iSWe0rdTz2A4W72CA,3732
|
163
178
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=256,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=5ro9O8Nf-7MB8NGCQ3QV5kB2k3iSWe0rdTz2A4W72CA,3732
|
@@ -168,8 +183,12 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=CYT3ujh5ifonhqQc1uYSa6maJ
|
|
168
183
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=Y4WpmVGlGERHoeoQNGkQ-GC-MsEtMblqnAVuDbARJdw,3240
|
169
184
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=64,device_name=NVIDIA_L20,dtype=int8_w8a8.json",sha256=RUkd9fW9WbajF_fFIzppsE1qyWGR5aRC4Cln-BPdu28,3254
|
170
185
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=64,device_name=NVIDIA_L40S,dtype=int8_w8a8.json",sha256=Sc9xK1wtRUqIzXppbutcq-Y2e9M0DZl2OGVzzB0aQuI,3265
|
171
|
-
"sglang/srt/layers/moe/fused_moe_triton/configs/E=
|
186
|
+
"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
|
187
|
+
"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
|
172
188
|
"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=0HPFNkhgQ_Yd0190i1bQSgth3q4zCfBgiRQJsITO-S8,3265
|
189
|
+
"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
|
190
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=272,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=Piw4LN6d8QYrUahWsw3XUOtTMD1o3vHPwA94sGI56Gk,3242
|
191
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=288,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=3T8_rF2PEojhgTMyQ8DscXgJCWWdWfDPj4M434zWcA4,3243
|
173
192
|
"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
|
174
193
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=64,N=1280,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=6QPLhZurIqcMVdy3w0Dd7gLViKxsyJRBz-qd8agpi6Q,3248
|
175
194
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=64,N=1280,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=uv-RwTNZT2n264dLo4eWxUpB3g7QqUyf2MFEGiRvoqQ,3251
|
@@ -243,25 +262,25 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=CYT3ujh5ifonhqQc1uYSa6maJ
|
|
243
262
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=8,N=8192,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8.json",sha256=-RzUWSIAAsg6iA-8SPMa68hPpBVoUyMJs3dLP7edRu0,4323
|
244
263
|
"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
|
245
264
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=8,N=8192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json",sha256=Uz5X80VcNBOaxshwVNUEittHk2zqB4HQCfTJ4TPG5aM,3274
|
246
|
-
sglang/srt/layers/quantization/__init__.py,sha256=
|
265
|
+
sglang/srt/layers/quantization/__init__.py,sha256=g5VWhdnQuyxFEcPyMhreR8QlEn3tXaCH34QPdAdufZQ,12369
|
247
266
|
sglang/srt/layers/quantization/awq.py,sha256=VImnVCU_QBLFba6S88T0dJ-vLy6SMm3OLIMEdllDfVI,6663
|
248
267
|
sglang/srt/layers/quantization/base_config.py,sha256=jWk_egQrVNMYmQgbTI9vkcgzScLFjB5_sywFlAfE5J0,4776
|
249
|
-
sglang/srt/layers/quantization/blockwise_int8.py,sha256=
|
250
|
-
sglang/srt/layers/quantization/fp8.py,sha256=
|
251
|
-
sglang/srt/layers/quantization/fp8_kernel.py,sha256=
|
252
|
-
sglang/srt/layers/quantization/fp8_utils.py,sha256=
|
268
|
+
sglang/srt/layers/quantization/blockwise_int8.py,sha256=yE8ARplbha1sW1Szl-mgsRDzGTRpEZY_zAKkCJIu680,15010
|
269
|
+
sglang/srt/layers/quantization/fp8.py,sha256=bYj6-xRO-bXsrDDaYzA2sKTmOvznLu7ZVoAtGR7cKjM,41834
|
270
|
+
sglang/srt/layers/quantization/fp8_kernel.py,sha256=F49gP48suKwzO1QejmGSV6XrBSOgwW-bsaM-rvUB_mE,27460
|
271
|
+
sglang/srt/layers/quantization/fp8_utils.py,sha256=ymdhxw-vMfJIzxW0uFg_iDTzlurN_R9dFXyEmYXsluI,22516
|
253
272
|
sglang/srt/layers/quantization/gptq.py,sha256=e4rMz374-yQQqeAI77WPxfcAaRk38GeN2akEpvnC_Do,15141
|
254
273
|
sglang/srt/layers/quantization/int8_kernel.py,sha256=GfRn_imIw8kNgqdtb2lr7BettjgDgimbl1Rubnamjh8,11352
|
255
274
|
sglang/srt/layers/quantization/int8_utils.py,sha256=YK9CS-lb_n91kNCTKK5o5apYF31V2giDg5G5VKrpcUA,2356
|
256
|
-
sglang/srt/layers/quantization/kv_cache.py,sha256
|
257
|
-
sglang/srt/layers/quantization/modelopt_quant.py,sha256=
|
258
|
-
sglang/srt/layers/quantization/moe_wna16.py,sha256=
|
275
|
+
sglang/srt/layers/quantization/kv_cache.py,sha256=-yaFTdB75T0BbvQeuIpH6rZoL3R8t6OIJVGB-xdtpCw,3492
|
276
|
+
sglang/srt/layers/quantization/modelopt_quant.py,sha256=Ff7qMv7CCWj0QY5gkDnwlQYLH1mbMtopbw8jXMLndXg,16616
|
277
|
+
sglang/srt/layers/quantization/moe_wna16.py,sha256=3Z8Eq4_ehTN5EEotlYC09FpUNmF8VO8uv7QzUqJa0QI,19371
|
259
278
|
sglang/srt/layers/quantization/utils.py,sha256=QqGFwRnFenOm5HfyLoS4D06_LyvNWgOggAiFtZXTpQ4,5637
|
260
|
-
sglang/srt/layers/quantization/w8a8_fp8.py,sha256=
|
261
|
-
sglang/srt/layers/quantization/w8a8_int8.py,sha256=
|
279
|
+
sglang/srt/layers/quantization/w8a8_fp8.py,sha256=wY9Ztw9RM5Vd3MHMLauy0KD8xcQ8JZUB_M4LeyE8-UU,11654
|
280
|
+
sglang/srt/layers/quantization/w8a8_int8.py,sha256=fIoSGeaL5kZNrExKAKWBgJc9hNXns_w9zP7vw2dVPHA,8892
|
262
281
|
sglang/srt/layers/quantization/compressed_tensors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
263
|
-
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=
|
264
|
-
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=
|
282
|
+
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=Z2h9WQ74Umun_0wj8I5fr-ScRfuSrpSU9otI29jLAM8,25606
|
283
|
+
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=RYX3vJ0Dh984KktTkNr1Zt578W9gp-YFY5nDiOpd7IQ,27073
|
265
284
|
sglang/srt/layers/quantization/compressed_tensors/utils.py,sha256=mnUmKWFQUnY8bVoFHUuNVwqsfS-cefeR-ofyaihCXcY,7621
|
266
285
|
sglang/srt/layers/quantization/compressed_tensors/schemes/__init__.py,sha256=qcdRgoUNJWXqSimns-D987TW0OTk3uFuWNBX8Z6H8Fk,246
|
267
286
|
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_scheme.py,sha256=tdKJC8c3SX8T3z8JL-1YCsg4ftcv55Wxt0vZrYftpX8,1635
|
@@ -438,66 +457,68 @@ sglang/srt/managers/configure_logging.py,sha256=fOJaXAQ1n9m-8KPJndpsKvS885i69SMa
|
|
438
457
|
sglang/srt/managers/data_parallel_controller.py,sha256=Xkj2n9uDyq7a-AVDZlfzeuNkC4ibsSftb1_bed9hgQ4,10318
|
439
458
|
sglang/srt/managers/detokenizer_manager.py,sha256=HTfpJWMF1EImhKOnLJ96xPmYXm71xzaisLMfxg3zpgs,10111
|
440
459
|
sglang/srt/managers/expert_distribution.py,sha256=r3o5RGI0gnV7xb60AApqKYa0oiSB37oB7hQBX7P3xZM,3225
|
441
|
-
sglang/srt/managers/io_struct.py,sha256=
|
442
|
-
sglang/srt/managers/mm_utils.py,sha256=
|
460
|
+
sglang/srt/managers/io_struct.py,sha256=88tlo4xLTqpb_qv2mJpJPZQxpcWCK3xW34GUij05McQ,30860
|
461
|
+
sglang/srt/managers/mm_utils.py,sha256=BdeiJG1vR89v8j9NwdHdanZlX5iiyxYWwHL5T0CjwUg,15537
|
443
462
|
sglang/srt/managers/multimodal_processor.py,sha256=37SSZIdhdmcGaZSH2A2GLdntcbIxDUiomX6WR_BpmtQ,2132
|
444
|
-
sglang/srt/managers/schedule_batch.py,sha256=
|
463
|
+
sglang/srt/managers/schedule_batch.py,sha256=4Xi7SdCiTKWraU8xQJQT1XYGMNt8TUW0dsZaNMDLutM,62509
|
445
464
|
sglang/srt/managers/schedule_policy.py,sha256=E1qVq2G3jptKdX9nlqfayeRBUll9xB6bK8nBf3EW32E,19469
|
446
|
-
sglang/srt/managers/scheduler.py,sha256=
|
465
|
+
sglang/srt/managers/scheduler.py,sha256=89db4YxUDZyrEZAKdbrxNIWk3EPob2aqvtZK4G8E2D0,82318
|
447
466
|
sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=u2sj6MViFTov0lVZSysZ-wph2pEqRCtCjwA1UdttZ7I,26338
|
448
467
|
sglang/srt/managers/session_controller.py,sha256=o-ifit0n4_xHLNmyD0Ams8FxGRgxFybX-Vz1hwgr3UQ,5755
|
449
|
-
sglang/srt/managers/tokenizer_manager.py,sha256=
|
450
|
-
sglang/srt/managers/tp_worker.py,sha256=
|
468
|
+
sglang/srt/managers/tokenizer_manager.py,sha256=ncPzDWeTgqi--V2LgRrzVIGp2aPl9Dcsv3qWCLDTBE4,46665
|
469
|
+
sglang/srt/managers/tp_worker.py,sha256=khF-hXOrtF_IesOyUSjEBjb7fAh3CakdiKR7Ebj2wp8,8894
|
451
470
|
sglang/srt/managers/tp_worker_overlap_thread.py,sha256=3_ZJ8Rq7v2ZDaRNTRu5Dy8AbqiAlJQp3IAKnn_WAwd8,9127
|
452
471
|
sglang/srt/managers/utils.py,sha256=5i75uLlQOF_5CaT02CrWtwozMTtwTg2_nLP8Dtr-JZQ,1536
|
453
|
-
sglang/srt/managers/multimodal_processors/base_processor.py,sha256=
|
472
|
+
sglang/srt/managers/multimodal_processors/base_processor.py,sha256=5pkKBqajyBRv7uM183NnrhTMYybC0HeUzHID_OkEnfA,9859
|
454
473
|
sglang/srt/managers/multimodal_processors/clip.py,sha256=lRc2mcuDbAhZVf-0EfkO81pqDiol9zLvTpDqtPIBQ2k,1525
|
455
474
|
sglang/srt/managers/multimodal_processors/deepseek_vl_v2.py,sha256=j7j1D38azudJjYthVpdz7jxQ9Z7SjwQfskpOIshAdiY,3147
|
456
475
|
sglang/srt/managers/multimodal_processors/gemma3.py,sha256=UlkyIoc8XOw69iFBYiBYLx--pdfnM4JfCFtwRrd3w-o,2267
|
457
|
-
sglang/srt/managers/multimodal_processors/janus_pro.py,sha256=
|
476
|
+
sglang/srt/managers/multimodal_processors/janus_pro.py,sha256=4bQZ7WByd53d1PcEgVeeXRyWnC78nO-8RsQbGWRDyYM,1852
|
458
477
|
sglang/srt/managers/multimodal_processors/llava.py,sha256=8mac3vUUpVd12o43k1TyMaLEySZB915ks8Q5epeZmbg,6209
|
459
478
|
sglang/srt/managers/multimodal_processors/minicpm.py,sha256=Mq-iH2j90VrGAbSaF3ayYWhTEm9RvWNI6ZhBb6G23dQ,5684
|
460
479
|
sglang/srt/managers/multimodal_processors/mlama.py,sha256=MLiGS606LzVtdoXvjWGANx-K_7nE9J_fMVmkXN7Gz8k,1661
|
480
|
+
sglang/srt/managers/multimodal_processors/mllama4.py,sha256=50Yox7TaGrrB7iPjN1dQ_UzuY41x7VLmMcRXBhTgUvE,5592
|
461
481
|
sglang/srt/managers/multimodal_processors/qwen_vl.py,sha256=67EmFiAkvZncU-eqiiS0Q4dr3pWcfI-RofYiQnNWvu0,5722
|
462
482
|
sglang/srt/mem_cache/base_prefix_cache.py,sha256=NY62Zo0A0tLJ7ObRLOQqQcXCxoJUDZsK8f5U4dNQjKc,973
|
463
483
|
sglang/srt/mem_cache/chunk_cache.py,sha256=it5SfL1FwMbrdeOH-I-Eu_i-I9hFB1xL-z_brIUoCkk,1835
|
464
484
|
sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
|
465
485
|
sglang/srt/mem_cache/hiradix_cache.py,sha256=t3zxOCg8A4uMdjrtKbSdDJBwFubqnhfGOEdSs_22Zb4,16161
|
466
|
-
sglang/srt/mem_cache/memory_pool.py,sha256=
|
486
|
+
sglang/srt/mem_cache/memory_pool.py,sha256=MyFVt81pCiiHoO3zRbIP_Z-KSAhHeSQGFQuL_kFL_L4,31638
|
467
487
|
sglang/srt/mem_cache/paged_allocator.py,sha256=BrJS0vN1k-vTSgb_M8u_1KoZFRgzgR1WRyImCTq3T0U,9770
|
468
488
|
sglang/srt/mem_cache/radix_cache.py,sha256=Lm-pco6CJ4orb9IfDpbHm5MnyK8Ya0OF1x9p88dv548,14906
|
469
|
-
sglang/srt/metrics/collector.py,sha256=
|
489
|
+
sglang/srt/metrics/collector.py,sha256=zHg4twFQJvuK1mSme3-EYQa9PJryfp_u7a4RxQ5RcO0,8874
|
470
490
|
sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
|
471
|
-
sglang/srt/model_executor/cuda_graph_runner.py,sha256=
|
472
|
-
sglang/srt/model_executor/forward_batch_info.py,sha256=
|
473
|
-
sglang/srt/model_executor/model_runner.py,sha256=
|
491
|
+
sglang/srt/model_executor/cuda_graph_runner.py,sha256=ulYmFv0mlQ4aawuFpX2qkaAVx3qE5tEYj4D7hOOEct8,23325
|
492
|
+
sglang/srt/model_executor/forward_batch_info.py,sha256=_qSMTiLxvcPIIgqRfUqG4W--OoirVY7ulcFfZqQIqjo,28689
|
493
|
+
sglang/srt/model_executor/model_runner.py,sha256=5qvLlql8rIMJUa8DTChrraq_7-s6PusnpdPctED3PJU,46909
|
474
494
|
sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
|
475
|
-
sglang/srt/model_loader/loader.py,sha256=
|
495
|
+
sglang/srt/model_loader/loader.py,sha256=YYmtvkQw0B1qgPw0_gN-K4yy7CEYbTSR__0Dl1Fnm6k,55342
|
476
496
|
sglang/srt/model_loader/utils.py,sha256=0NaMR67fESFopaklmsleiL27XH1QUrjZW246MUu1EJ0,1369
|
477
|
-
sglang/srt/model_loader/weight_utils.py,sha256=
|
478
|
-
sglang/srt/models/baichuan.py,sha256=
|
479
|
-
sglang/srt/models/chatglm.py,sha256=
|
497
|
+
sglang/srt/model_loader/weight_utils.py,sha256=yKnau-wH9muczoCpDTCVIqXFqz-QJmEEySplX3bMJWk,32153
|
498
|
+
sglang/srt/models/baichuan.py,sha256=HbvlErnkCSK4pRQYCSDxMcrn-1DQyfiNoeDcnRrJas8,15807
|
499
|
+
sglang/srt/models/chatglm.py,sha256=cajLN9caBl09e0TwOFkiTTKDqwlbmHo_yS-NCjdeQW8,13957
|
480
500
|
sglang/srt/models/clip.py,sha256=fCMtAcaKjruSIWfD4YGb4HXh6Tzp2pjpgDmp5JpwBPU,19794
|
481
|
-
sglang/srt/models/commandr.py,sha256=
|
482
|
-
sglang/srt/models/dbrx.py,sha256=
|
483
|
-
sglang/srt/models/deepseek.py,sha256=
|
501
|
+
sglang/srt/models/commandr.py,sha256=5Y_b3K0QY7D37nFGkyiGgY38RleRui_GJUYcHSuHUZo,15315
|
502
|
+
sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,15598
|
503
|
+
sglang/srt/models/deepseek.py,sha256=jZFQUVJ753qcI8_3sh6TlLF-8oYiQndQm-3No8FInXk,16910
|
484
504
|
sglang/srt/models/deepseek_janus_pro.py,sha256=8wAzvcGdyo--3faMN4QtagT1eAZMhMFduvpCXqUS48Q,70456
|
485
505
|
sglang/srt/models/deepseek_nextn.py,sha256=kca-2Fm2_SmqbOEFfd80pobooi1BXd1oe_4EsUM6SeI,13561
|
486
|
-
sglang/srt/models/deepseek_v2.py,sha256=
|
506
|
+
sglang/srt/models/deepseek_v2.py,sha256=LtaJOxEn6ZnsXkiiVdS646u7lw68TeBg3iPy2LGj_cY,70572
|
487
507
|
sglang/srt/models/deepseek_vl2.py,sha256=RVvi_3qsfrkqMCCnjjTA8OwUc5ySutc7asAH-rUJLVo,12922
|
488
|
-
sglang/srt/models/exaone.py,sha256=
|
489
|
-
sglang/srt/models/gemma.py,sha256=
|
490
|
-
sglang/srt/models/gemma2.py,sha256=
|
508
|
+
sglang/srt/models/exaone.py,sha256=rX7J0xFt9TSt6tMIhnYMkb5KDnqTJIV4BtjPLFwQ8_8,13425
|
509
|
+
sglang/srt/models/gemma.py,sha256=4cdrPISg1VKnsuI-QPTpYvet4BrX8BMKvCIN82iLskw,12641
|
510
|
+
sglang/srt/models/gemma2.py,sha256=kqtwdo93GWKm2iBN29RoIRH2ggRm-K_80LM5btgfBLo,16395
|
491
511
|
sglang/srt/models/gemma2_reward.py,sha256=V8U3_ADUHWPdOwvEe1jhGW-oJmBgL8t1TY3-67Ksv2A,2618
|
492
|
-
sglang/srt/models/gemma3_causal.py,sha256=
|
512
|
+
sglang/srt/models/gemma3_causal.py,sha256=G_vNxBZBJOVigZg8B3XGtbSmak9LEtSFFr9uQMXBHQY,24973
|
493
513
|
sglang/srt/models/gemma3_mm.py,sha256=tWX2vIdRf5zePwKMLbb0d24DUWoTdjmdXnxIcULQJ2E,15221
|
494
|
-
sglang/srt/models/gpt2.py,sha256=
|
495
|
-
sglang/srt/models/gpt_bigcode.py,sha256=
|
496
|
-
sglang/srt/models/granite.py,sha256=
|
497
|
-
sglang/srt/models/grok.py,sha256=
|
498
|
-
sglang/srt/models/internlm2.py,sha256=
|
514
|
+
sglang/srt/models/gpt2.py,sha256=kclhxEs8oJk1KCyhmAqo7rZqecVGGHYkc-a1WZi3aIk,9841
|
515
|
+
sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtYhl4,10307
|
516
|
+
sglang/srt/models/granite.py,sha256=5WOJyNYAlt5RNHSexNfPNihhSxIMd7wPzju1cTixKig,20852
|
517
|
+
sglang/srt/models/grok.py,sha256=vESZeGS4adI_JAerXIkCcTm15-CNiGeS7VHc36C6w1A,28033
|
518
|
+
sglang/srt/models/internlm2.py,sha256=RDAT9drjdgVEFmCMq99RTn3weMQFhl1NHhkhyDX8f7M,13056
|
499
519
|
sglang/srt/models/internlm2_reward.py,sha256=ndfGmyqYZbVZ7C7rJ-v9oK3wa-EpoBGybS8MlyKZi2E,2522
|
500
|
-
sglang/srt/models/llama.py,sha256=
|
520
|
+
sglang/srt/models/llama.py,sha256=71GmA-_-CNM2kuEJplNg6tfWbjCW31EzkeVIk5ZwNmo,24932
|
521
|
+
sglang/srt/models/llama4.py,sha256=JIVS5Q1lnmEpAHDI487gKO_9xfTCehSpzInNQeCg8JU,17940
|
501
522
|
sglang/srt/models/llama_classification.py,sha256=4QWTFaUZIFKYZvEzs8bx8VkOZNIwdYCLrnwrdAw4QK0,3108
|
502
523
|
sglang/srt/models/llama_eagle.py,sha256=OB2lKsjn7BcfCZljklnhk83me8j0PuQmYLou7baNcq4,4866
|
503
524
|
sglang/srt/models/llama_eagle3.py,sha256=v3bftBVDIGjnzngQYnu19cy0J_3w7yruHqLP5nsAQDM,6642
|
@@ -505,33 +526,34 @@ sglang/srt/models/llama_embedding.py,sha256=zq-_lNu35VBFc7eemiam0zdkGIE8fzrgk5OW
|
|
505
526
|
sglang/srt/models/llama_reward.py,sha256=LF2nqMV5XOrljGjAwJg43mBv3z6Q040I2EYlgZeCp8k,4681
|
506
527
|
sglang/srt/models/llava.py,sha256=KMwNNrlMuMaKEOZMDRBKBQbe6uctpKTLc0zOceyGC34,27242
|
507
528
|
sglang/srt/models/llavavid.py,sha256=q0lHlRnoYHKJZsWnkIQdd6dYAQ26t7XsmrqA0zDGmZc,12829
|
508
|
-
sglang/srt/models/minicpm.py,sha256
|
509
|
-
sglang/srt/models/minicpm3.py,sha256=
|
529
|
+
sglang/srt/models/minicpm.py,sha256=m5HFsSJj0Po09LY9R6qj6K4gceqWDMOePz3NDGgMGT4,14691
|
530
|
+
sglang/srt/models/minicpm3.py,sha256=ZQpk6j2UjtVDR5gA0_jGYvl5Vsvm7NBH7xkpNjqgGw0,26348
|
510
531
|
sglang/srt/models/minicpmo.py,sha256=kJnp8UwJTV7kXEpuVWA50ecRsuZyFedHlwkprix8tag,75619
|
511
532
|
sglang/srt/models/minicpmv.py,sha256=79zZn3co9r7SERatx49EuHRoLWRiy6qeaUFgjDWJo2I,40571
|
512
533
|
sglang/srt/models/mistral.py,sha256=EYifJUUzN2Z2-iL37eJiNZF_DB0H4pa0mKlgYRIxM70,838
|
513
|
-
sglang/srt/models/mixtral.py,sha256=
|
514
|
-
sglang/srt/models/mixtral_quant.py,sha256
|
515
|
-
sglang/srt/models/mllama.py,sha256=
|
516
|
-
sglang/srt/models/
|
517
|
-
sglang/srt/models/
|
518
|
-
sglang/srt/models/
|
519
|
-
sglang/srt/models/
|
520
|
-
sglang/srt/models/
|
521
|
-
sglang/srt/models/
|
522
|
-
sglang/srt/models/
|
534
|
+
sglang/srt/models/mixtral.py,sha256=zQHCL_ZMKmLR7jitpEw8n7Rv6xhxUJzSXklsw6auh2E,14965
|
535
|
+
sglang/srt/models/mixtral_quant.py,sha256=-kQw9r8KcLdO8SNN9RKXzrGq9Q2Al9l9cWHi1VrZSRM,15260
|
536
|
+
sglang/srt/models/mllama.py,sha256=jYV5ckyuJN5XU2VXjUgV1i-Yz5rZDQ-6OYsNZvUTJjo,39775
|
537
|
+
sglang/srt/models/mllama4.py,sha256=65_YDBaNSeJJuigz3sZKvsq25ZGO17MQIoya8ukJgRA,9086
|
538
|
+
sglang/srt/models/olmo.py,sha256=7-q_fA6XXdG7kPUjpUzYkzMUWJobuSjhqjYw9xSUs_c,12671
|
539
|
+
sglang/srt/models/olmo2.py,sha256=azmljhJF4ivcQfUtfsAUxq3ducE4tRKTL6iwe0IKYMg,14327
|
540
|
+
sglang/srt/models/olmoe.py,sha256=TMzt-yB891bvA4X50xL0NjNnFYSx9imlA7N1EG8KNK0,15949
|
541
|
+
sglang/srt/models/phi3_small.py,sha256=UbqZvpwWolXUPd0zbKgbL93yVXUY1n4kXJLgIe_gjaM,15508
|
542
|
+
sglang/srt/models/qwen.py,sha256=xYkVmMZS2uMqWhfndc8EYm0olpKFnggfuMp_6aobVi4,10758
|
543
|
+
sglang/srt/models/qwen2.py,sha256=fYE5fkyRYTEVVl8XnQO2-ybj4ZhNtM7Kn12AQt39EDA,16181
|
544
|
+
sglang/srt/models/qwen2_5_vl.py,sha256=uNnYhY8x-9H1GzUJkj7lUtR5d-0yMRNWUcT7-4qPlMU,22555
|
523
545
|
sglang/srt/models/qwen2_classification.py,sha256=dGrMm4ebd30_lBhHOhaV57ig2iOTx3nqB4GEzsrRIM8,2747
|
524
546
|
sglang/srt/models/qwen2_eagle.py,sha256=Iz0HWL2FgSD3FqoFhfYmbIZeEYkPTJ96lYbkncmHJX4,4644
|
525
|
-
sglang/srt/models/qwen2_moe.py,sha256=
|
547
|
+
sglang/srt/models/qwen2_moe.py,sha256=GhDR7pP_G0NZ2HkaFVrBZnbqB0RxxNnH-8HMLwrweE4,18245
|
526
548
|
sglang/srt/models/qwen2_rm.py,sha256=-mQXDEv11p-I1HXgYLTtY6ROem6UYorO958WsDrzsgs,2837
|
527
|
-
sglang/srt/models/qwen2_vl.py,sha256=
|
549
|
+
sglang/srt/models/qwen2_vl.py,sha256=NCG85isoPkepv5RU-eLh44rCHPhfT3bu7pifNdBEsVw,21612
|
528
550
|
sglang/srt/models/registry.py,sha256=inKh9iwOp3LFYm3nqujg-OtABClOP-ifc1stA9cZegA,3434
|
529
|
-
sglang/srt/models/stablelm.py,sha256=
|
551
|
+
sglang/srt/models/stablelm.py,sha256=0x_31uIr3WcWwecdPAI3ek9KkyKBJS7VwknTk2y0gjY,12281
|
530
552
|
sglang/srt/models/torch_native_llama.py,sha256=5tfFSMAXB3ScToqTALtCXa8Oo-qPCJh-KQCNB6QOlNA,19293
|
531
|
-
sglang/srt/models/xverse.py,sha256=
|
532
|
-
sglang/srt/models/xverse_moe.py,sha256=
|
553
|
+
sglang/srt/models/xverse.py,sha256=DsNVI9JpzN4jj0Ry6aTrj7r-xq5YLOoDX2kH4YLJA-I,14035
|
554
|
+
sglang/srt/models/xverse_moe.py,sha256=7KCM2-j12towDMNvXkuuYiBOmNauH6NG4Ip40x0khqA,16782
|
533
555
|
sglang/srt/models/yivl.py,sha256=oToK7-u5IGO7xwpJIQ7VtudlK6-zPqJX4bt6_wv0SH8,4850
|
534
|
-
sglang/srt/openai_api/adapter.py,sha256=
|
556
|
+
sglang/srt/openai_api/adapter.py,sha256=DaSU4Pri70s3ZeMHeVzsnKjd8dA9lx_HOmpVs1TEepo,72095
|
535
557
|
sglang/srt/openai_api/protocol.py,sha256=Y8PFFhLbzhpoERM6-WsTkm-ZuGcE-3tfenh9e-AC1vc,13374
|
536
558
|
sglang/srt/platforms/interface.py,sha256=hym3iooBB4C8if5hDZezgVN6h4NIOu7sg2ZUBIV6XmM,11246
|
537
559
|
sglang/srt/sampling/custom_logit_processor.py,sha256=tDvoLgLqn-sy1qcY6vSrpbnHCeqbdk0uhMOO-uy4p4E,1099
|
@@ -543,9 +565,9 @@ sglang/srt/sampling/penaltylib/min_new_tokens.py,sha256=rdU_D7RoIcrQPhysNQEzmr4T
|
|
543
565
|
sglang/srt/sampling/penaltylib/orchestrator.py,sha256=XM-Lm6u7gYPtMZrTIc0FR4QxNZxBH5s_Cj82umyCzYk,5721
|
544
566
|
sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=NRh10AJrrQlGJ6S-enGdRefrTrWpyqrSm-aNnyqQNQQ,2119
|
545
567
|
sglang/srt/speculative/build_eagle_tree.py,sha256=SFQ3eHbhfNxOdxgqDP5wSV_ZlIVqLw7VivycNZ963N0,11690
|
546
|
-
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=
|
547
|
-
sglang/srt/speculative/eagle_utils.py,sha256=
|
548
|
-
sglang/srt/speculative/eagle_worker.py,sha256=
|
568
|
+
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=FP-Dc6K4zaL2KQA8QsNccBM8TXnwREh1I2iPL9KHo8I,9252
|
569
|
+
sglang/srt/speculative/eagle_utils.py,sha256=mv--nBUgAbqP30pU3aGEMwQIHBwwevETUMQSZAelApE,28721
|
570
|
+
sglang/srt/speculative/eagle_worker.py,sha256=Qvg3B40GKH6vjyrZ9SmiVyW6KbuRJJHYXbnRCvSz3aE,27016
|
549
571
|
sglang/srt/speculative/spec_info.py,sha256=rhaKG0TzyF9XZYHEWp1jccwTBohSNsUDvxHFtAoOl18,709
|
550
572
|
sglang/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
551
573
|
sglang/test/few_shot_gsm8k.py,sha256=7VLbWl4nCQs1wjtW4q-46jf9jUCycSs5Iw8v7sUSzBw,4284
|
@@ -560,17 +582,19 @@ sglang/test/simple_eval_math.py,sha256=6kGKNwNbLN-Af3Wj8WTimWhH-Xp3enDmSvvSjsgWU
|
|
560
582
|
sglang/test/simple_eval_mgsm.py,sha256=rd7TSUyxdKbrXaVoewo24V8lCo_6kO8zxPhhmvylpw8,10259
|
561
583
|
sglang/test/simple_eval_mmlu.py,sha256=FkwamjGMjueTixymkedF-YiPloSLiy4ftILFUrKZ9XI,4357
|
562
584
|
sglang/test/test_activation.py,sha256=GeTIJHxlLQfW3kM-X1FGa8Sa3dSGKHEXl5wEy-hfGis,1489
|
563
|
-
sglang/test/test_block_fp8.py,sha256=
|
585
|
+
sglang/test/test_block_fp8.py,sha256=6Ux1_E6EWdY184n8tiYOCwbyHVAUEqz9lMhSUDLIOC8,16292
|
564
586
|
sglang/test/test_block_fp8_ep.py,sha256=N1rvqbPErBaFFpeAw8TLYXGNZOoG7cfIBP2p5XbSyMo,10806
|
565
587
|
sglang/test/test_custom_ops.py,sha256=4X3-odkJntwNtBAuKtCbYHu6peIP6LaI_VwLw7kmDx8,5550
|
566
588
|
sglang/test/test_dynamic_grad_mode.py,sha256=L76yUCuk_ymNpXD2CmO8r2GiGjIvD_gtTsuFDs2NolI,1638
|
567
589
|
sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg,3764
|
568
590
|
sglang/test/test_programs.py,sha256=VZ3vXtUDBnXz0M7gFdDH8hXg9Wa0j_qI8CVqjEgRN_E,18877
|
569
|
-
sglang/test/test_utils.py,sha256=
|
591
|
+
sglang/test/test_utils.py,sha256=Y7XMx8-BTQJr6a90qRVpK4x9Lkl_p2WyL0VwFNHxhPs,30530
|
570
592
|
sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
571
|
-
sglang/test/attention/test_flashattn_backend.py,sha256=
|
572
|
-
sglang
|
573
|
-
sglang
|
574
|
-
sglang-0.4.
|
575
|
-
sglang-0.4.
|
576
|
-
sglang-0.4.
|
593
|
+
sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
|
594
|
+
sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
|
595
|
+
sglang/test/attention/test_prefix_chunk_info.py,sha256=er0i3KGHMkw-4UZB1GCFd4oYwRcXfU5wpO1ORqpNGGA,7626
|
596
|
+
sglang-0.4.5.post1.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
|
597
|
+
sglang-0.4.5.post1.dist-info/METADATA,sha256=602BdUHYnIS2M5riEnyqkhTpEKDRbH0J0ubdy1FK8fg,25571
|
598
|
+
sglang-0.4.5.post1.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
|
599
|
+
sglang-0.4.5.post1.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
|
600
|
+
sglang-0.4.5.post1.dist-info/RECORD,,
|
@@ -1,81 +0,0 @@
|
|
1
|
-
from __future__ import annotations
|
2
|
-
|
3
|
-
import logging
|
4
|
-
from enum import Enum
|
5
|
-
from typing import Optional
|
6
|
-
|
7
|
-
import numpy as np
|
8
|
-
import numpy.typing as npt
|
9
|
-
|
10
|
-
logger = logging.getLogger(__name__)
|
11
|
-
|
12
|
-
|
13
|
-
class KVArgs:
|
14
|
-
engine_rank: int
|
15
|
-
kv_data_ptrs: list[int]
|
16
|
-
kv_data_lens: list[int]
|
17
|
-
kv_item_lens: list[int]
|
18
|
-
aux_data_ptrs: list[int]
|
19
|
-
aux_data_lens: list[int]
|
20
|
-
aux_item_lens: list[int]
|
21
|
-
ib_device: str
|
22
|
-
|
23
|
-
|
24
|
-
class KVManager:
|
25
|
-
def __init__(self, args: KVArgs): ...
|
26
|
-
|
27
|
-
|
28
|
-
class KVPoll:
|
29
|
-
Failed = 0
|
30
|
-
Bootstrapping = 1
|
31
|
-
WaitingForInput = 2
|
32
|
-
Transferring = 3
|
33
|
-
Success = 4
|
34
|
-
|
35
|
-
|
36
|
-
class KVSender:
|
37
|
-
def __init__(self, mgr: KVManager, bootstrap_addr: str, bootstrap_room: int):
|
38
|
-
self.has_sent = False
|
39
|
-
|
40
|
-
def init(self, num_kv_indices: int, aux_index: Optional[int] = None): ...
|
41
|
-
|
42
|
-
def send(self, kv_indices: npt.NDArray[np.int32]):
|
43
|
-
self.has_sent = True
|
44
|
-
|
45
|
-
def poll(self) -> KVPoll:
|
46
|
-
if self.has_sent is False:
|
47
|
-
# Assume handshake completed instantly
|
48
|
-
return KVPoll.WaitingForInput
|
49
|
-
else:
|
50
|
-
# Assume transfer completed instantly
|
51
|
-
return KVPoll.Success
|
52
|
-
|
53
|
-
def failure_exception(self):
|
54
|
-
raise Exception("Fake KVSender Exception")
|
55
|
-
|
56
|
-
|
57
|
-
class KVReceiver:
|
58
|
-
def __init__(
|
59
|
-
self, mgr: KVManager, bootstrap_addr: str, bootstrap_room: Optional[int] = None
|
60
|
-
):
|
61
|
-
self.has_init = False
|
62
|
-
|
63
|
-
def init(self, kv_indices: npt.NDArray[np.int32], aux_index: Optional[int] = None):
|
64
|
-
self.has_init = True
|
65
|
-
|
66
|
-
def poll(self) -> KVPoll:
|
67
|
-
if self.has_init is False:
|
68
|
-
# Assume handshake completed instantly
|
69
|
-
return KVPoll.WaitingForInput
|
70
|
-
else:
|
71
|
-
# Assume transfer completed instantly
|
72
|
-
return KVPoll.Success
|
73
|
-
|
74
|
-
def failure_exception(self):
|
75
|
-
raise Exception("Fake KVReceiver Exception")
|
76
|
-
|
77
|
-
|
78
|
-
class KVBootstrapServer:
|
79
|
-
def __init__(self, port: int): ...
|
80
|
-
|
81
|
-
def poll(self) -> KVPoll: ...
|
File without changes
|
File without changes
|
File without changes
|