sglang 0.4.4.post2__py3-none-any.whl → 0.4.4.post4__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_serving.py +72 -10
- sglang/srt/_custom_ops.py +59 -92
- sglang/srt/configs/deepseekvl2.py +10 -1
- sglang/srt/configs/model_config.py +6 -16
- sglang/srt/constrained/base_grammar_backend.py +5 -1
- sglang/srt/custom_op.py +5 -0
- sglang/srt/distributed/device_communicators/custom_all_reduce.py +28 -80
- sglang/srt/distributed/device_communicators/custom_all_reduce_utils.py +2 -2
- sglang/srt/distributed/parallel_state.py +32 -5
- sglang/srt/entrypoints/engine.py +0 -5
- sglang/srt/entrypoints/http_server.py +7 -1
- sglang/srt/entrypoints/verl_engine.py +2 -0
- sglang/srt/function_call_parser.py +0 -1
- sglang/srt/layers/attention/flashattention_backend.py +582 -125
- sglang/srt/layers/attention/flashinfer_backend.py +5 -7
- sglang/srt/layers/attention/flashinfer_mla_backend.py +1 -3
- sglang/srt/layers/attention/flashmla_backend.py +1 -1
- sglang/srt/layers/dp_attention.py +12 -1
- sglang/srt/layers/moe/ep_moe/kernels.py +142 -0
- sglang/srt/layers/moe/ep_moe/layer.py +79 -80
- sglang/srt/layers/moe/ep_moe/token_dispatcher.py +382 -199
- sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_H20,block_shape=[128, 128].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 +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/E=264,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +403 -47
- sglang/srt/layers/moe/topk.py +79 -6
- sglang/srt/layers/quantization/__init__.py +137 -165
- sglang/srt/layers/quantization/awq.py +200 -0
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py +2 -1
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py +34 -10
- sglang/srt/layers/quantization/fp8_kernel.py +2 -1
- sglang/srt/layers/quantization/fp8_utils.py +1 -4
- sglang/srt/layers/quantization/gptq.py +30 -40
- sglang/srt/layers/quantization/moe_wna16.py +501 -0
- sglang/srt/layers/quantization/utils.py +1 -1
- sglang/srt/layers/quantization/w8a8_fp8.py +1 -1
- sglang/srt/lora/backend/base_backend.py +4 -4
- sglang/srt/lora/backend/flashinfer_backend.py +12 -9
- sglang/srt/lora/backend/triton_backend.py +5 -8
- sglang/srt/lora/layers.py +19 -33
- sglang/srt/lora/lora_manager.py +20 -7
- sglang/srt/lora/mem_pool.py +12 -6
- sglang/srt/lora/triton_ops/gate_up_lora_b.py +10 -4
- sglang/srt/lora/triton_ops/qkv_lora_b.py +8 -3
- sglang/srt/lora/triton_ops/sgemm_lora_a.py +16 -5
- sglang/srt/lora/triton_ops/sgemm_lora_b.py +11 -6
- sglang/srt/lora/utils.py +6 -0
- sglang/srt/managers/cache_controller.py +34 -11
- sglang/srt/managers/io_struct.py +4 -2
- sglang/srt/managers/mm_utils.py +202 -156
- sglang/srt/managers/multimodal_processor.py +0 -2
- sglang/srt/managers/multimodal_processors/base_processor.py +45 -77
- sglang/srt/managers/multimodal_processors/clip.py +44 -0
- sglang/srt/managers/multimodal_processors/deepseek_vl_v2.py +17 -58
- sglang/srt/managers/multimodal_processors/gemma3.py +12 -27
- sglang/srt/managers/multimodal_processors/janus_pro.py +21 -47
- sglang/srt/managers/multimodal_processors/llava.py +34 -14
- sglang/srt/managers/multimodal_processors/minicpm.py +35 -38
- sglang/srt/managers/multimodal_processors/mlama.py +10 -23
- sglang/srt/managers/multimodal_processors/qwen_vl.py +22 -45
- sglang/srt/managers/schedule_batch.py +185 -127
- sglang/srt/managers/scheduler.py +29 -23
- sglang/srt/managers/tokenizer_manager.py +1 -2
- sglang/srt/managers/tp_worker.py +3 -0
- sglang/srt/managers/utils.py +1 -6
- sglang/srt/mem_cache/hiradix_cache.py +62 -52
- sglang/srt/mem_cache/memory_pool.py +72 -6
- sglang/srt/mem_cache/paged_allocator.py +39 -0
- sglang/srt/metrics/collector.py +23 -53
- sglang/srt/model_executor/cuda_graph_runner.py +16 -13
- sglang/srt/model_executor/forward_batch_info.py +10 -10
- sglang/srt/model_executor/model_runner.py +64 -59
- sglang/srt/model_loader/loader.py +19 -1
- sglang/srt/model_loader/weight_utils.py +6 -3
- sglang/srt/models/clip.py +568 -0
- sglang/srt/models/deepseek_janus_pro.py +12 -17
- sglang/srt/models/deepseek_v2.py +339 -123
- sglang/srt/models/deepseek_vl2.py +105 -104
- sglang/srt/models/gemma3_causal.py +12 -2
- sglang/srt/models/gemma3_mm.py +20 -80
- sglang/srt/models/llama.py +4 -1
- sglang/srt/models/llava.py +31 -19
- sglang/srt/models/llavavid.py +16 -7
- sglang/srt/models/minicpmo.py +63 -147
- sglang/srt/models/minicpmv.py +17 -27
- sglang/srt/models/mllama.py +29 -14
- sglang/srt/models/qwen2.py +9 -6
- sglang/srt/models/qwen2_5_vl.py +21 -31
- sglang/srt/models/qwen2_vl.py +20 -21
- sglang/srt/openai_api/adapter.py +106 -93
- sglang/srt/openai_api/protocol.py +10 -5
- sglang/srt/patch_torch.py +71 -0
- sglang/srt/platforms/interface.py +371 -0
- sglang/srt/server_args.py +120 -25
- sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +5 -5
- sglang/srt/speculative/eagle_utils.py +140 -28
- sglang/srt/speculative/eagle_worker.py +94 -25
- sglang/srt/utils.py +137 -51
- sglang/test/runners.py +27 -2
- sglang/test/test_custom_ops.py +55 -0
- sglang/test/test_utils.py +14 -27
- sglang/utils.py +2 -2
- sglang/version.py +1 -1
- {sglang-0.4.4.post2.dist-info → sglang-0.4.4.post4.dist-info}/METADATA +10 -5
- {sglang-0.4.4.post2.dist-info → sglang-0.4.4.post4.dist-info}/RECORD +108 -99
- {sglang-0.4.4.post2.dist-info → sglang-0.4.4.post4.dist-info}/WHEEL +0 -0
- {sglang-0.4.4.post2.dist-info → sglang-0.4.4.post4.dist-info}/licenses/LICENSE +0 -0
- {sglang-0.4.4.post2.dist-info → sglang-0.4.4.post4.dist-info}/top_level.txt +0 -0
@@ -3,13 +3,13 @@ sglang/api.py,sha256=vHiKBg8wwIdmrpnGclop5BzJ-1Q88emrlrfLwNCHg98,7010
|
|
3
3
|
sglang/bench_offline_throughput.py,sha256=OQb-AjL4UNymmir02ht43uzgaNsnO_I11nXSowKMqBI,13841
|
4
4
|
sglang/bench_one_batch.py,sha256=Fp6HBBJHrw672Q1gnklJ7dYboYYjR92D2fNCvbrM3M0,17935
|
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=DKCg7l1uaDlKUB45AIpFaZLesA-sRTV-meJ-50sucXE,57410
|
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
|
-
sglang/utils.py,sha256=
|
12
|
-
sglang/version.py,sha256=
|
11
|
+
sglang/utils.py,sha256=GIcgiRHkZ-gyPxXOdn1qFF41jkg4-YdDxbPc4mzO-qk,16159
|
12
|
+
sglang/version.py,sha256=63MgDRjbuREW1iqTD8A1g5qw4uGX6I-V8Zgt5c1nnys,28
|
13
13
|
sglang/lang/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
14
14
|
sglang/lang/chat_template.py,sha256=xZ-kQpgb4-NY6QOqgRjOODZW5G4EvJnplaqYF6Ng2Ow,18952
|
15
15
|
sglang/lang/choices.py,sha256=-W1DVw9N9ZliVpvmWrzIXG4cswAah8eMQrHWzkS3D8o,6234
|
@@ -24,30 +24,31 @@ sglang/lang/backend/litellm.py,sha256=ugmL7sfUxkUHVbHtwNzHgdQAEd4UCjNQboFuE3KThc
|
|
24
24
|
sglang/lang/backend/openai.py,sha256=6I1udBC9obVpJG0GerbFVbWCbwLKkF2DoPdHBCnqPzs,16341
|
25
25
|
sglang/lang/backend/runtime_endpoint.py,sha256=CAVh3X9F80t_2tkJECF__7AdCQtqDg1AHDqIoKIPnvs,16755
|
26
26
|
sglang/lang/backend/vertexai.py,sha256=O-iBLD-y3vq80UxnrAoJri7bxpgd-_eakZ88Cf8bEGA,4855
|
27
|
-
sglang/srt/_custom_ops.py,sha256=
|
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
30
|
sglang/srt/conversation.py,sha256=mzNPZX7ndgmm1E8azuK8eytN6bNCEu3WbcPReRFfhd0,27819
|
31
|
-
sglang/srt/custom_op.py,sha256=
|
32
|
-
sglang/srt/function_call_parser.py,sha256=
|
31
|
+
sglang/srt/custom_op.py,sha256=bIZ__3FiZvkbsN9O_jeLy_49X7ZbYbw0VxoL80uWwaI,3715
|
32
|
+
sglang/srt/function_call_parser.py,sha256=buYENeNEP5bhsvD424yGCa9wOqSfVOZSRn6zLiSJp5I,23733
|
33
33
|
sglang/srt/hf_transformers_utils.py,sha256=_QYTl9LpU0jmKPlYooHi1etwMvb5v40JIrG_t_Fx06w,9215
|
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=Fw_QGqSsAdyCumi0dT2HyPlppf9xd3-tQPvwuBZfhxU,2625
|
36
37
|
sglang/srt/reasoning_parser.py,sha256=45xsU9RCPfyG4_Zx4y3-JPyNgAtrqwKI4j5R2NT4g1s,5594
|
37
38
|
sglang/srt/server.py,sha256=PrQb9r6L9syWHKlggbbiQYsKtpwSmECqozRbf8qnoV8,874
|
38
|
-
sglang/srt/server_args.py,sha256=
|
39
|
+
sglang/srt/server_args.py,sha256=eb3zJIpljzHK_ajp_zJRgwRUM_00-S-7K15k2opaBK4,52467
|
39
40
|
sglang/srt/torch_memory_saver_adapter.py,sha256=KG3wM9-xZsSdsmORofArnNR7hH55GEyFxaderCDcK9w,1853
|
40
|
-
sglang/srt/utils.py,sha256=
|
41
|
+
sglang/srt/utils.py,sha256=UyNimlcXkBG5cp-6ah3GaBBTUCpOA-OZh0NaNNh7QgA,57507
|
41
42
|
sglang/srt/warmup.py,sha256=FmJiYfjRr3X_eAe7ojQaPoN17LvHpjDmRWRnO-k86AQ,1469
|
42
43
|
sglang/srt/configs/__init__.py,sha256=vulncVn70WqIT6s0HaB8p_Q6FjOiaLwNZWpoJS9FIuQ,399
|
43
44
|
sglang/srt/configs/chatglm.py,sha256=j-b0YkdYUmQm2y1kNmMJtKeACxWKmBbvNNkDWbs6kbI,2907
|
44
45
|
sglang/srt/configs/dbrx.py,sha256=tdhIkXAQl1yr0MxqFmsDG1E0e2puRTTKm6UTyANBLac,11005
|
45
|
-
sglang/srt/configs/deepseekvl2.py,sha256=
|
46
|
+
sglang/srt/configs/deepseekvl2.py,sha256=qeI7TO4f2DHIxCRiuRxpnP17K5lCMSNGJjEcq2yF9Ds,23328
|
46
47
|
sglang/srt/configs/device_config.py,sha256=kfmpPOECqYxcRoY-ko0QZRhyiBWUGP2CMF51DMUN5nU,435
|
47
48
|
sglang/srt/configs/exaone.py,sha256=Duxd4yQoKy8GWEzZD_kCY_OzmN_67CTJL_Kgn0eXk3g,10731
|
48
49
|
sglang/srt/configs/janus_pro.py,sha256=-QtJ4ZGZiAJb0AkOEcuCHzIKLw23nF8nRk3rdCcoUO0,19016
|
49
50
|
sglang/srt/configs/load_config.py,sha256=qs-AxuplouBx2tsv9KGBOLZPbwzuVA4vbktbGP_cRp8,3309
|
50
|
-
sglang/srt/configs/model_config.py,sha256
|
51
|
+
sglang/srt/configs/model_config.py,sha256=-6SEpQw-U_KWONdNYqEe8nXopqDcDtp4B4g-RZQPhHo,20581
|
51
52
|
sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
|
52
53
|
sglang/srt/connector/__init__.py,sha256=czLX5JOxuMhH-T9eSJzoc1qv1B4z9chyffDRL5I6wo4,1247
|
53
54
|
sglang/srt/connector/base_connector.py,sha256=i6i1TIzsz4NbSEkrdMPq-urb2sN2aLAx8dazga4gB9U,2833
|
@@ -57,7 +58,7 @@ sglang/srt/connector/utils.py,sha256=isTvxauz1_8no5MW6p8Bwj2H9mQkweaRO_RSwAPA7R8
|
|
57
58
|
sglang/srt/connector/serde/__init__.py,sha256=MvBJ7FBJtnou-AD-EdsCdAIDCcA8kWvUNuRViILVJ-0,718
|
58
59
|
sglang/srt/connector/serde/safe_serde.py,sha256=Iv_mFsDvk-cXuw0WTykBZ2yGxW0jb82DwVqC7r3RmAU,750
|
59
60
|
sglang/srt/connector/serde/serde.py,sha256=n59I2MXLa7WCyN_8pEd8L-scJk7lMhmEX-GOUIhF0ZA,1004
|
60
|
-
sglang/srt/constrained/base_grammar_backend.py,sha256=
|
61
|
+
sglang/srt/constrained/base_grammar_backend.py,sha256=MzAA7oqWOZ12ndUs158FGECjtKNx_2_mDMZ7Jopb9Pk,6899
|
61
62
|
sglang/srt/constrained/llguidance_backend.py,sha256=ej7wN13SzCsT310C6OIyUg2zs5jeuLl3Ocok9SP9-c4,5702
|
62
63
|
sglang/srt/constrained/outlines_backend.py,sha256=UWv2xjg8x4XtoqpY8LoorlJaYOZhfDeIr5YCiFn4knA,6812
|
63
64
|
sglang/srt/constrained/outlines_jump_forward.py,sha256=iZWXeR3gNYoMubLGyFmLPO4V2YsN5DiGjD71Xk9iFaE,6418
|
@@ -69,21 +70,21 @@ sglang/srt/disaggregation/prefill.py,sha256=zw8hDy6Txq_MpC5j0fndLNcKoypT2BhxTkqq
|
|
69
70
|
sglang/srt/disaggregation/utils.py,sha256=ebOZ3lSFVkbNtl6uUfS6sYYYVBjgmWdQLOsqIZBGgN0,1088
|
70
71
|
sglang/srt/distributed/__init__.py,sha256=jFOcyt-wFAPMBUAf9zkZalNQlt-4rqmT6pCKBz1E4qo,149
|
71
72
|
sglang/srt/distributed/communication_op.py,sha256=IBnFUdMftK_VSTMMMitGveonorFUUVNL4guqO31cMSc,1130
|
72
|
-
sglang/srt/distributed/parallel_state.py,sha256=
|
73
|
+
sglang/srt/distributed/parallel_state.py,sha256=hoTgLYfHIKMb_tSwBTauuusJZ8oY9BsiubTTOF8UfIw,50713
|
73
74
|
sglang/srt/distributed/utils.py,sha256=U-BSaXYjWwnfG8g-tUfBhjKt5Ug097nyHtu3g3aea_Y,8473
|
74
75
|
sglang/srt/distributed/device_communicators/cuda_wrapper.py,sha256=3jvPG-Ow5UBLiXhfx8T8snR7crSZbPpARAggsDPWq7k,7038
|
75
|
-
sglang/srt/distributed/device_communicators/custom_all_reduce.py,sha256=
|
76
|
-
sglang/srt/distributed/device_communicators/custom_all_reduce_utils.py,sha256=
|
76
|
+
sglang/srt/distributed/device_communicators/custom_all_reduce.py,sha256=OClh322wSV28K_LpUyXX2SiasAFh7yZr6vPDG84rj9o,19913
|
77
|
+
sglang/srt/distributed/device_communicators/custom_all_reduce_utils.py,sha256=fLoptT_U0lVAqkhEg-ge53CdFSIKQpDRiqHYKwJVEZg,10974
|
77
78
|
sglang/srt/distributed/device_communicators/hpu_communicator.py,sha256=gPjEH1-izoby5uDrfUlzNf21luPT0Ow7pJjhCRKnHy8,1728
|
78
79
|
sglang/srt/distributed/device_communicators/pynccl.py,sha256=G-Dut_QJHOUG0j7--ZqapHtvm70Lgl7obtE6ZfgeAiU,10064
|
79
80
|
sglang/srt/distributed/device_communicators/pynccl_wrapper.py,sha256=LblisImY9d6EMz-oPS9J16WHo2Q_SRL1DtlJKK63Hfg,15349
|
80
81
|
sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=bbruDIM1GgKIdB6gi71_I0mpB179I-qyvwKuSj1Kaic,20816
|
81
82
|
sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
|
82
|
-
sglang/srt/entrypoints/engine.py,sha256=
|
83
|
-
sglang/srt/entrypoints/http_server.py,sha256=
|
84
|
-
sglang/srt/entrypoints/verl_engine.py,sha256=
|
83
|
+
sglang/srt/entrypoints/engine.py,sha256=1ML85N-iF22n2Id2dpbYrKxxfkqnVUXP26kEGUfcA2E,21146
|
84
|
+
sglang/srt/entrypoints/http_server.py,sha256=VM-gVwW-Ef_SikxoVSE06yydlLo6TGJykjKgffnGoXY,27104
|
85
|
+
sglang/srt/entrypoints/verl_engine.py,sha256=PypBCkUJYy7iP3mKB-W0KYkjZzs4Rq6DqxNgG_nBZaM,5903
|
85
86
|
sglang/srt/layers/activation.py,sha256=1ykXZO0BGz7DFVE-EK26b02I5AgH2IuU4PQB6oUcF4M,6003
|
86
|
-
sglang/srt/layers/dp_attention.py,sha256=
|
87
|
+
sglang/srt/layers/dp_attention.py,sha256=fC1kaYkHGoFjZ2KHTzPFW0e93El-XLRt7ZidkXYIVhE,7595
|
87
88
|
sglang/srt/layers/elementwise.py,sha256=y2mQqjbF2FmFtNYBk5ecTyaj3ELoZyz-rWPY8rrxCtk,13765
|
88
89
|
sglang/srt/layers/layernorm.py,sha256=189bORMggKhYcEYEjl6JRcuIoUPllHo3SheoH6YiORY,4546
|
89
90
|
sglang/srt/layers/linear.py,sha256=HYIGxpRYL6x-jNOkyNtGAw5Ak9Nq8jkntddgTBER_1w,51486
|
@@ -97,10 +98,10 @@ sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSa
|
|
97
98
|
sglang/srt/layers/vocab_parallel_embedding.py,sha256=QUxd4sELx6p3dHvEKmccPZ-phdd_9EjNdwjH3SJ9zxI,22238
|
98
99
|
sglang/srt/layers/attention/base_attn_backend.py,sha256=X_GIbQuU9njtUEGdUP7E_KRhmGxj3UyPHNESlL3QaQ8,3264
|
99
100
|
sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
|
100
|
-
sglang/srt/layers/attention/flashattention_backend.py,sha256=
|
101
|
-
sglang/srt/layers/attention/flashinfer_backend.py,sha256=
|
102
|
-
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=
|
103
|
-
sglang/srt/layers/attention/flashmla_backend.py,sha256=
|
101
|
+
sglang/srt/layers/attention/flashattention_backend.py,sha256=IwrJRmBFWLUdYqe7n4Dd2rX4zu9V55qBLjCdNQEmkd0,30306
|
102
|
+
sglang/srt/layers/attention/flashinfer_backend.py,sha256=3fxS2NQzCBw7h_gLxBjHcyDkf2quWqBxr_N01lYmfJo,45865
|
103
|
+
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=pnVhvVEK87iFW8gUb1G7X7c1tqro8R2DSEOFCnlV8Bo,30301
|
104
|
+
sglang/srt/layers/attention/flashmla_backend.py,sha256=1RPFNtQOBw6BWxIjrzfJgA9Nx92udLbR-S5KXmqjxS8,10536
|
104
105
|
sglang/srt/layers/attention/torch_native_backend.py,sha256=KABmBrMqKa4x08kkQYdIcZUGydvmaVJIUfo3y8jhFHI,9270
|
105
106
|
sglang/srt/layers/attention/triton_backend.py,sha256=cyxOaUU1CNhaEezJH9j0dd20cwxwIVGGN3jNXFTVkIY,26714
|
106
107
|
sglang/srt/layers/attention/utils.py,sha256=J9mA-cbZT3uTlaKXo0HEAaeMei_TS2o4McTna9LVDCE,2750
|
@@ -112,13 +113,13 @@ sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=Y66gZ37u0GKMP
|
|
112
113
|
sglang/srt/layers/attention/triton_ops/rocm_mla_decode_rope.py,sha256=664WnAJ91EiCUZOcnVDfbTQf4uGJ4ZDZB1CbxpEUFZc,13866
|
113
114
|
sglang/srt/layers/moe/fused_moe_native.py,sha256=KF0J5W5biWYWbERUHAduQFnUV5om9VaYv0I9avsVfgU,4330
|
114
115
|
sglang/srt/layers/moe/router.py,sha256=gvyK7hXlujfCZCmAIFc3oxfgjuAjzlpPe3mp1Blc6Y0,10419
|
115
|
-
sglang/srt/layers/moe/topk.py,sha256=
|
116
|
+
sglang/srt/layers/moe/topk.py,sha256=iUb-64CaNAUfvBZ1pkgsedcLRQs2sVSIzQ5300WmdXI,10242
|
116
117
|
sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
117
|
-
sglang/srt/layers/moe/ep_moe/kernels.py,sha256=
|
118
|
-
sglang/srt/layers/moe/ep_moe/layer.py,sha256=
|
119
|
-
sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=
|
118
|
+
sglang/srt/layers/moe/ep_moe/kernels.py,sha256=ijqRzS-tb0LGnDU5hW-g0JH104ppADrWaUIDGxb9Feo,22919
|
119
|
+
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=jnr6KSM8YooftTjZ3gYe0eWpOd1dmkXqk4hKRvLTwCo,19708
|
120
121
|
sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=h9yMFAL_bagUf-qBED8gSWdCOb7d8IdA-pE-L_nIg8E,842
|
121
|
-
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=
|
122
|
+
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=_ufAA7m6alppU4vVxh7QrgSgFOcBSekLQ-UQdHq7TmQ,53988
|
122
123
|
sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=CYT3ujh5ifonhqQc1uYSa6maJADUOBJoRHMIWmf-DU4,24100
|
123
124
|
"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
|
124
125
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=JJN0hryyLr5Zv3dSS7C8cPFhAwTT6XxUVnBGMZvV6JA,2752
|
@@ -156,6 +157,7 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=CYT3ujh5ifonhqQc1uYSa6maJ
|
|
156
157
|
"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
|
157
158
|
"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
|
158
159
|
"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
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=128,device_name=NVIDIA_H20,block_shape=[128, 128].json",sha256=JEOXj48phwoumZWBwNq9TpqxVwIrbnfot4QfAdzvLJI,3249
|
159
161
|
"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
|
160
162
|
"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
|
161
163
|
"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
|
@@ -166,6 +168,8 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=CYT3ujh5ifonhqQc1uYSa6maJ
|
|
166
168
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=256,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=Y4WpmVGlGERHoeoQNGkQ-GC-MsEtMblqnAVuDbARJdw,3240
|
167
169
|
"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
|
168
170
|
"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=257,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=z8Iedw3N22cWXTCPhVBBk-yZqkc30ePMzv9KkgPoOd0,3266
|
172
|
+
"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
|
169
173
|
"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
|
170
174
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=64,N=1280,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=6QPLhZurIqcMVdy3w0Dd7gLViKxsyJRBz-qd8agpi6Q,3248
|
171
175
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=64,N=1280,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=uv-RwTNZT2n264dLo4eWxUpB3g7QqUyf2MFEGiRvoqQ,3251
|
@@ -239,23 +243,25 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=CYT3ujh5ifonhqQc1uYSa6maJ
|
|
239
243
|
"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
|
240
244
|
"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
|
241
245
|
"sglang/srt/layers/moe/fused_moe_triton/configs/E=8,N=8192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json",sha256=Uz5X80VcNBOaxshwVNUEittHk2zqB4HQCfTJ4TPG5aM,3274
|
242
|
-
sglang/srt/layers/quantization/__init__.py,sha256=
|
246
|
+
sglang/srt/layers/quantization/__init__.py,sha256=TxMVIyn2PiP_xrZQugOmzrfZ7DEKEjdNOuFhDNBWBys,12107
|
247
|
+
sglang/srt/layers/quantization/awq.py,sha256=VImnVCU_QBLFba6S88T0dJ-vLy6SMm3OLIMEdllDfVI,6663
|
243
248
|
sglang/srt/layers/quantization/base_config.py,sha256=jWk_egQrVNMYmQgbTI9vkcgzScLFjB5_sywFlAfE5J0,4776
|
244
249
|
sglang/srt/layers/quantization/blockwise_int8.py,sha256=BS4nmo18QaC3vyCe9Wwe0Te-7FmFUe-udl7xbhRaU0s,14887
|
245
250
|
sglang/srt/layers/quantization/fp8.py,sha256=J9P6SwZ1PIb3dEJoP7X2Uw5VOtU35vU5jugkoYyaZC4,41081
|
246
|
-
sglang/srt/layers/quantization/fp8_kernel.py,sha256=
|
247
|
-
sglang/srt/layers/quantization/fp8_utils.py,sha256=
|
248
|
-
sglang/srt/layers/quantization/gptq.py,sha256=
|
251
|
+
sglang/srt/layers/quantization/fp8_kernel.py,sha256=JRalHJ-btDpzl3oXu2R_ZoJBu5TzBBmW_wKZDFs-usQ,24384
|
252
|
+
sglang/srt/layers/quantization/fp8_utils.py,sha256=CDR2fLrZa_mZ86n5S2dDjYMpVCGa2n7gCXd2BYZjXcM,21391
|
253
|
+
sglang/srt/layers/quantization/gptq.py,sha256=e4rMz374-yQQqeAI77WPxfcAaRk38GeN2akEpvnC_Do,15141
|
249
254
|
sglang/srt/layers/quantization/int8_kernel.py,sha256=GfRn_imIw8kNgqdtb2lr7BettjgDgimbl1Rubnamjh8,11352
|
250
255
|
sglang/srt/layers/quantization/int8_utils.py,sha256=YK9CS-lb_n91kNCTKK5o5apYF31V2giDg5G5VKrpcUA,2356
|
251
256
|
sglang/srt/layers/quantization/kv_cache.py,sha256=rJi6amyLZsquUMo_V5iLlPMqdsGTLgxh4popN1xUHCQ,4236
|
252
257
|
sglang/srt/layers/quantization/modelopt_quant.py,sha256=mne4uKF0R-K0OvWN7X5ZxD4LdXKBc6GvmpZzIW6gkmM,6969
|
253
|
-
sglang/srt/layers/quantization/
|
254
|
-
sglang/srt/layers/quantization/
|
258
|
+
sglang/srt/layers/quantization/moe_wna16.py,sha256=JlYCiy825MEMlBeQqqWwPuhoH_FbEcgTYlJUzrWj5IE,19248
|
259
|
+
sglang/srt/layers/quantization/utils.py,sha256=QqGFwRnFenOm5HfyLoS4D06_LyvNWgOggAiFtZXTpQ4,5637
|
260
|
+
sglang/srt/layers/quantization/w8a8_fp8.py,sha256=XcQdgqXA3eKbAf-4_0I81Y5Nvjns3bQTocovnN8141w,6234
|
255
261
|
sglang/srt/layers/quantization/w8a8_int8.py,sha256=V5vxn0wmUL1szj38lsJOKeNNEvFHisU7hZZLO4FfoNc,8733
|
256
262
|
sglang/srt/layers/quantization/compressed_tensors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
257
|
-
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=
|
258
|
-
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=
|
263
|
+
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=ngKVSHfQUNSZzrLMu4Iv_4Fzt2eOoOIZKcO2RNDiwAM,25353
|
264
|
+
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=roqRrIJybA9YuN3kqSeoLTJhXfTHOOtJd5MkenpOL8E,25835
|
259
265
|
sglang/srt/layers/quantization/compressed_tensors/utils.py,sha256=mnUmKWFQUnY8bVoFHUuNVwqsfS-cefeR-ofyaihCXcY,7621
|
260
266
|
sglang/srt/layers/quantization/compressed_tensors/schemes/__init__.py,sha256=qcdRgoUNJWXqSimns-D987TW0OTk3uFuWNBX8Z6H8Fk,246
|
261
267
|
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_scheme.py,sha256=tdKJC8c3SX8T3z8JL-1YCsg4ftcv55Wxt0vZrYftpX8,1635
|
@@ -412,119 +418,122 @@ sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a
|
|
412
418
|
"sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=PD4AJYCkHfy2ivv9baMouFXzBTy0eKMumbAfxfm91HI,3256
|
413
419
|
"sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json",sha256=FImA-TJ_tQDjqwoNWxS--sRDoKDXf9gamlME3tkxH58,3252
|
414
420
|
"sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FFBjSWlpKXMxfAUUYUqXbOK_Hd7qBeBsfbcaa9uB4qY,3249
|
415
|
-
sglang/srt/lora/layers.py,sha256=
|
421
|
+
sglang/srt/lora/layers.py,sha256=eqU5PxLx9jsmp0fxQ-e9mlSD0Zz4Y9Uan_x9Z5-y1EQ,11835
|
416
422
|
sglang/srt/lora/lora.py,sha256=9BpasJObx-XL_qSoQhheGaYBJ3PlKqJAARvruIOJE0w,7514
|
417
423
|
sglang/srt/lora/lora_config.py,sha256=qDgMTx_69jyJUl29O5FxLzYa0BMhqYVXWXfyyVOvGm0,1684
|
418
|
-
sglang/srt/lora/lora_manager.py,sha256=
|
419
|
-
sglang/srt/lora/mem_pool.py,sha256=
|
420
|
-
sglang/srt/lora/utils.py,sha256=
|
424
|
+
sglang/srt/lora/lora_manager.py,sha256=Gpkq4N_cJGMIDtxUCScwP4LGcHyUJZ457EI_ti30_A8,9187
|
425
|
+
sglang/srt/lora/mem_pool.py,sha256=xUFoHUDJgX9lt2YugD9HUY5tIMnJiazYMZ6LYqSGv-E,9633
|
426
|
+
sglang/srt/lora/utils.py,sha256=GjEBgsGhDhX4NqVqeaciznQ8RotKZmb2c-nw4YMLHxA,5251
|
421
427
|
sglang/srt/lora/backend/__init__.py,sha256=FziFT8HguMFj-h0tUCc4_UEbtOWMlYi4gNlYJcArWh4,671
|
422
|
-
sglang/srt/lora/backend/base_backend.py,sha256=
|
423
|
-
sglang/srt/lora/backend/flashinfer_backend.py,sha256=
|
424
|
-
sglang/srt/lora/backend/triton_backend.py,sha256=
|
428
|
+
sglang/srt/lora/backend/base_backend.py,sha256=tGpABAn3DVC8GONf8USkaxkzkpVsDYfgKrnLCsXpivo,4558
|
429
|
+
sglang/srt/lora/backend/flashinfer_backend.py,sha256=VmDSY2YqTLK2EBeqdMiNoirPxDifCMmfiCB3HNwpgvE,4138
|
430
|
+
sglang/srt/lora/backend/triton_backend.py,sha256=_QbqggFPHMPWgx3PI15yyyfxBCTxSpGA209x_frI12E,2517
|
425
431
|
sglang/srt/lora/triton_ops/__init__.py,sha256=JGOYPIn1XbGcyJTbt8A0qoc02PYONSGNNjGkC8yJpAM,283
|
426
|
-
sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=
|
427
|
-
sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=
|
428
|
-
sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=
|
429
|
-
sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=
|
430
|
-
sglang/srt/managers/cache_controller.py,sha256=
|
432
|
+
sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=CDGt7lpu9GjykgMtmwbZ3PEqjTlRYyh28AUlj1cRcmw,5279
|
433
|
+
sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=HTfU3HxxxVyaG_aJrrVjPJTnqf62yvepcKJKYkG0XJQ,5944
|
434
|
+
sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=ZmWEqHJaorRNNj-c_ZXPi_pX8X_yIAwudRHAJVa0m08,4350
|
435
|
+
sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=Q58UzWUb3QFqY_ZxWA3poN373N0Hwkks5AQRKIuvFC8,4517
|
436
|
+
sglang/srt/managers/cache_controller.py,sha256=d4RGqbut1FlzJnpqr7WY_TYmRjYPS07OoOVbztjs5xI,18959
|
431
437
|
sglang/srt/managers/configure_logging.py,sha256=fOJaXAQ1n9m-8KPJndpsKvS885i69SMafoEADLIVfIM,1633
|
432
438
|
sglang/srt/managers/data_parallel_controller.py,sha256=Xkj2n9uDyq7a-AVDZlfzeuNkC4ibsSftb1_bed9hgQ4,10318
|
433
439
|
sglang/srt/managers/detokenizer_manager.py,sha256=HTfpJWMF1EImhKOnLJ96xPmYXm71xzaisLMfxg3zpgs,10111
|
434
440
|
sglang/srt/managers/expert_distribution.py,sha256=r3o5RGI0gnV7xb60AApqKYa0oiSB37oB7hQBX7P3xZM,3225
|
435
|
-
sglang/srt/managers/io_struct.py,sha256=
|
436
|
-
sglang/srt/managers/mm_utils.py,sha256=
|
437
|
-
sglang/srt/managers/multimodal_processor.py,sha256=
|
438
|
-
sglang/srt/managers/schedule_batch.py,sha256=
|
441
|
+
sglang/srt/managers/io_struct.py,sha256=_WYQ2x49Wc8WqmZH0Q5Te7zVrGRQkbn0ADghuDwyk7k,23852
|
442
|
+
sglang/srt/managers/mm_utils.py,sha256=KxZF684q0ohUn4J4dPMdDfGtOKLyWUZ3o7yG-mGcjnY,15464
|
443
|
+
sglang/srt/managers/multimodal_processor.py,sha256=37SSZIdhdmcGaZSH2A2GLdntcbIxDUiomX6WR_BpmtQ,2132
|
444
|
+
sglang/srt/managers/schedule_batch.py,sha256=oHXIHW8imrDnV0PuOjysG9qvEVYqQY9XGCTMKzDd-6I,61013
|
439
445
|
sglang/srt/managers/schedule_policy.py,sha256=E1qVq2G3jptKdX9nlqfayeRBUll9xB6bK8nBf3EW32E,19469
|
440
|
-
sglang/srt/managers/scheduler.py,sha256=
|
446
|
+
sglang/srt/managers/scheduler.py,sha256=hSYFlzkr20ZCYVvPfffmmG_aQeLUx9xjDCcvudEnU3Y,80024
|
441
447
|
sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=u2sj6MViFTov0lVZSysZ-wph2pEqRCtCjwA1UdttZ7I,26338
|
442
448
|
sglang/srt/managers/session_controller.py,sha256=o-ifit0n4_xHLNmyD0Ams8FxGRgxFybX-Vz1hwgr3UQ,5755
|
443
|
-
sglang/srt/managers/tokenizer_manager.py,sha256=
|
444
|
-
sglang/srt/managers/tp_worker.py,sha256=
|
449
|
+
sglang/srt/managers/tokenizer_manager.py,sha256=dSuYrkMNZUcI1hSNP4sw5fHV4FdndysNeRG1NsxBEjo,46163
|
450
|
+
sglang/srt/managers/tp_worker.py,sha256=IFiOhbNIya-7cqgp_Yg0ZXGcsgy9YS295AfxJYjFqzQ,8833
|
445
451
|
sglang/srt/managers/tp_worker_overlap_thread.py,sha256=3_ZJ8Rq7v2ZDaRNTRu5Dy8AbqiAlJQp3IAKnn_WAwd8,9127
|
446
|
-
sglang/srt/managers/utils.py,sha256=
|
447
|
-
sglang/srt/managers/multimodal_processors/base_processor.py,sha256=
|
448
|
-
sglang/srt/managers/multimodal_processors/
|
449
|
-
sglang/srt/managers/multimodal_processors/
|
450
|
-
sglang/srt/managers/multimodal_processors/
|
451
|
-
sglang/srt/managers/multimodal_processors/
|
452
|
-
sglang/srt/managers/multimodal_processors/
|
453
|
-
sglang/srt/managers/multimodal_processors/
|
454
|
-
sglang/srt/managers/multimodal_processors/
|
452
|
+
sglang/srt/managers/utils.py,sha256=5i75uLlQOF_5CaT02CrWtwozMTtwTg2_nLP8Dtr-JZQ,1536
|
453
|
+
sglang/srt/managers/multimodal_processors/base_processor.py,sha256=8ELm-cEJgFnOh8DxzsgRlCEGjewA68IygJRZGlF8Azw,8923
|
454
|
+
sglang/srt/managers/multimodal_processors/clip.py,sha256=lRc2mcuDbAhZVf-0EfkO81pqDiol9zLvTpDqtPIBQ2k,1525
|
455
|
+
sglang/srt/managers/multimodal_processors/deepseek_vl_v2.py,sha256=j7j1D38azudJjYthVpdz7jxQ9Z7SjwQfskpOIshAdiY,3147
|
456
|
+
sglang/srt/managers/multimodal_processors/gemma3.py,sha256=UlkyIoc8XOw69iFBYiBYLx--pdfnM4JfCFtwRrd3w-o,2267
|
457
|
+
sglang/srt/managers/multimodal_processors/janus_pro.py,sha256=wZs4HZhPov7yvV2VU2ep4k1ANOimVqPRIs3cpC-O1I4,1820
|
458
|
+
sglang/srt/managers/multimodal_processors/llava.py,sha256=8mac3vUUpVd12o43k1TyMaLEySZB915ks8Q5epeZmbg,6209
|
459
|
+
sglang/srt/managers/multimodal_processors/minicpm.py,sha256=Mq-iH2j90VrGAbSaF3ayYWhTEm9RvWNI6ZhBb6G23dQ,5684
|
460
|
+
sglang/srt/managers/multimodal_processors/mlama.py,sha256=MLiGS606LzVtdoXvjWGANx-K_7nE9J_fMVmkXN7Gz8k,1661
|
461
|
+
sglang/srt/managers/multimodal_processors/qwen_vl.py,sha256=67EmFiAkvZncU-eqiiS0Q4dr3pWcfI-RofYiQnNWvu0,5722
|
455
462
|
sglang/srt/mem_cache/base_prefix_cache.py,sha256=NY62Zo0A0tLJ7ObRLOQqQcXCxoJUDZsK8f5U4dNQjKc,973
|
456
463
|
sglang/srt/mem_cache/chunk_cache.py,sha256=it5SfL1FwMbrdeOH-I-Eu_i-I9hFB1xL-z_brIUoCkk,1835
|
457
464
|
sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
|
458
|
-
sglang/srt/mem_cache/hiradix_cache.py,sha256=
|
459
|
-
sglang/srt/mem_cache/memory_pool.py,sha256=
|
460
|
-
sglang/srt/mem_cache/paged_allocator.py,sha256=
|
465
|
+
sglang/srt/mem_cache/hiradix_cache.py,sha256=t3zxOCg8A4uMdjrtKbSdDJBwFubqnhfGOEdSs_22Zb4,16161
|
466
|
+
sglang/srt/mem_cache/memory_pool.py,sha256=v5Mzx1VuyRpZ7P3liiKKfXuw_t24kjYPpqV_ZCwGCeA,31098
|
467
|
+
sglang/srt/mem_cache/paged_allocator.py,sha256=BrJS0vN1k-vTSgb_M8u_1KoZFRgzgR1WRyImCTq3T0U,9770
|
461
468
|
sglang/srt/mem_cache/radix_cache.py,sha256=Lm-pco6CJ4orb9IfDpbHm5MnyK8Ya0OF1x9p88dv548,14906
|
462
|
-
sglang/srt/metrics/collector.py,sha256=
|
469
|
+
sglang/srt/metrics/collector.py,sha256=aCxHqgsQ6P8ZxsAvq_MoEVsr3KUvIUSOBpGYMgBxmOM,8442
|
463
470
|
sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
|
464
|
-
sglang/srt/model_executor/cuda_graph_runner.py,sha256=
|
465
|
-
sglang/srt/model_executor/forward_batch_info.py,sha256=
|
466
|
-
sglang/srt/model_executor/model_runner.py,sha256=
|
471
|
+
sglang/srt/model_executor/cuda_graph_runner.py,sha256=bDLOqlxdwRUyKitG8JyZygnm05N00q-TdNiAayG_T8o,23223
|
472
|
+
sglang/srt/model_executor/forward_batch_info.py,sha256=8VI1VxSmyH26lIHnCNeGqYw2XxslbqN_cuSUIEPUtRU,19468
|
473
|
+
sglang/srt/model_executor/model_runner.py,sha256=f2e0Ts7eTy5_xCIp7q2If4Mi6Oi3nw45zyBPvNgR_-0,45006
|
467
474
|
sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
|
468
|
-
sglang/srt/model_loader/loader.py,sha256=
|
475
|
+
sglang/srt/model_loader/loader.py,sha256=AUS4SqSFghbQjs29C65lg7_zxR9h1t7N5G0gERjc0Rc,54238
|
469
476
|
sglang/srt/model_loader/utils.py,sha256=0NaMR67fESFopaklmsleiL27XH1QUrjZW246MUu1EJ0,1369
|
470
|
-
sglang/srt/model_loader/weight_utils.py,sha256=
|
477
|
+
sglang/srt/model_loader/weight_utils.py,sha256=Bkfgz6TUjkJJb8BiFxrv7FgbZFd9eW2y21jDBPdLWSo,32056
|
471
478
|
sglang/srt/models/baichuan.py,sha256=iXgta-W38OWesxmXWZJ73fUvPdu51EwTQzUD5mmfJ8s,15721
|
472
479
|
sglang/srt/models/chatglm.py,sha256=avLC7mjjGskBxCxy-9s0sMlAJjfFoG_y8VieR1QfDsM,13918
|
480
|
+
sglang/srt/models/clip.py,sha256=fCMtAcaKjruSIWfD4YGb4HXh6Tzp2pjpgDmp5JpwBPU,19794
|
473
481
|
sglang/srt/models/commandr.py,sha256=Ug-B0QcdWZufrTybC6K5yP3MLKNsYb-vzfrqUsXYUcI,15276
|
474
482
|
sglang/srt/models/dbrx.py,sha256=0Vf4yhqe8YeQuKR3P-agvYJScmHwH3-tFbyU8kv5QJM,15559
|
475
483
|
sglang/srt/models/deepseek.py,sha256=Le2MXij8m4hT7QYgD0bFMFmYhbgX7SMjoXZFB8BxgyA,16871
|
476
|
-
sglang/srt/models/deepseek_janus_pro.py,sha256
|
484
|
+
sglang/srt/models/deepseek_janus_pro.py,sha256=8wAzvcGdyo--3faMN4QtagT1eAZMhMFduvpCXqUS48Q,70456
|
477
485
|
sglang/srt/models/deepseek_nextn.py,sha256=kca-2Fm2_SmqbOEFfd80pobooi1BXd1oe_4EsUM6SeI,13561
|
478
|
-
sglang/srt/models/deepseek_v2.py,sha256=
|
479
|
-
sglang/srt/models/deepseek_vl2.py,sha256=
|
486
|
+
sglang/srt/models/deepseek_v2.py,sha256=HJ8cuH87E_EF62YXlmYBjGVJk5P721T5M-XKMqsrbYg,62633
|
487
|
+
sglang/srt/models/deepseek_vl2.py,sha256=RVvi_3qsfrkqMCCnjjTA8OwUc5ySutc7asAH-rUJLVo,12922
|
480
488
|
sglang/srt/models/exaone.py,sha256=5iibqQTjpgosuGRt2rj2lWR0ShK2XGhbdFSnOWpaQss,13386
|
481
489
|
sglang/srt/models/gemma.py,sha256=3XxMDOKz4xMP6VzWoW8f0hmMf8LP8fhzMw5prsYC4e8,12602
|
482
490
|
sglang/srt/models/gemma2.py,sha256=MDe_HNkSpEJpw426tbx3fp271GBlSVEuhIdGeOB_jYA,16356
|
483
491
|
sglang/srt/models/gemma2_reward.py,sha256=V8U3_ADUHWPdOwvEe1jhGW-oJmBgL8t1TY3-67Ksv2A,2618
|
484
|
-
sglang/srt/models/gemma3_causal.py,sha256=
|
485
|
-
sglang/srt/models/gemma3_mm.py,sha256=
|
492
|
+
sglang/srt/models/gemma3_causal.py,sha256=nKO-DRtvcXn5bHquxILgnp0fJT6usoB81W8kPVdYsA0,24934
|
493
|
+
sglang/srt/models/gemma3_mm.py,sha256=tWX2vIdRf5zePwKMLbb0d24DUWoTdjmdXnxIcULQJ2E,15221
|
486
494
|
sglang/srt/models/gpt2.py,sha256=dAnfmsAL7JVHakryqrERR1jgL8mI1Op6nPHYfDCF7Ao,9802
|
487
495
|
sglang/srt/models/gpt_bigcode.py,sha256=EAN6xAXpa8m3DcBuH1D4rTPji2oG9NSozGXSNHtE2lw,10268
|
488
496
|
sglang/srt/models/granite.py,sha256=nu_Zl_PYn188gk1uYVZ76y4wwHZV7G0w7uanhqpSFUs,20813
|
489
497
|
sglang/srt/models/grok.py,sha256=pQOXtpHOYVntwt5QQRLffYsnMHmMfPMmGyKMfR0k0Ic,27994
|
490
498
|
sglang/srt/models/internlm2.py,sha256=4eh9WVgK4yg13IsnH5qB2xUCWnixj_aLLz7qa_4m2_Q,13017
|
491
499
|
sglang/srt/models/internlm2_reward.py,sha256=ndfGmyqYZbVZ7C7rJ-v9oK3wa-EpoBGybS8MlyKZi2E,2522
|
492
|
-
sglang/srt/models/llama.py,sha256=
|
500
|
+
sglang/srt/models/llama.py,sha256=FRdD-oCQl1M4hkYwIrWuECQ1tgxBq9rEIYOr-rlgswE,24616
|
493
501
|
sglang/srt/models/llama_classification.py,sha256=4QWTFaUZIFKYZvEzs8bx8VkOZNIwdYCLrnwrdAw4QK0,3108
|
494
502
|
sglang/srt/models/llama_eagle.py,sha256=OB2lKsjn7BcfCZljklnhk83me8j0PuQmYLou7baNcq4,4866
|
495
503
|
sglang/srt/models/llama_eagle3.py,sha256=v3bftBVDIGjnzngQYnu19cy0J_3w7yruHqLP5nsAQDM,6642
|
496
504
|
sglang/srt/models/llama_embedding.py,sha256=zq-_lNu35VBFc7eemiam0zdkGIE8fzrgk5OWYfirZnA,3254
|
497
505
|
sglang/srt/models/llama_reward.py,sha256=LF2nqMV5XOrljGjAwJg43mBv3z6Q040I2EYlgZeCp8k,4681
|
498
|
-
sglang/srt/models/llava.py,sha256=
|
499
|
-
sglang/srt/models/llavavid.py,sha256=
|
506
|
+
sglang/srt/models/llava.py,sha256=KMwNNrlMuMaKEOZMDRBKBQbe6uctpKTLc0zOceyGC34,27242
|
507
|
+
sglang/srt/models/llavavid.py,sha256=q0lHlRnoYHKJZsWnkIQdd6dYAQ26t7XsmrqA0zDGmZc,12829
|
500
508
|
sglang/srt/models/minicpm.py,sha256=-ot45U_Bv4x85JdbIAQXoxa1sF-ZDkBk8flU-Ruli5Y,14652
|
501
509
|
sglang/srt/models/minicpm3.py,sha256=sRHPFUH636GIY94B-hpAN2MSzYT1pzLPVypTNjUtttY,26270
|
502
|
-
sglang/srt/models/minicpmo.py,sha256=
|
503
|
-
sglang/srt/models/minicpmv.py,sha256=
|
510
|
+
sglang/srt/models/minicpmo.py,sha256=kJnp8UwJTV7kXEpuVWA50ecRsuZyFedHlwkprix8tag,75619
|
511
|
+
sglang/srt/models/minicpmv.py,sha256=79zZn3co9r7SERatx49EuHRoLWRiy6qeaUFgjDWJo2I,40571
|
504
512
|
sglang/srt/models/mistral.py,sha256=EYifJUUzN2Z2-iL37eJiNZF_DB0H4pa0mKlgYRIxM70,838
|
505
513
|
sglang/srt/models/mixtral.py,sha256=6Fse2J-20IMylP-yzpEihIinaH37TmmslATbLcWBRYY,14926
|
506
514
|
sglang/srt/models/mixtral_quant.py,sha256=MSa6UKPbgv8Rn8Iv8o1dQhcstAHLNQzE0eepFx_hYSw,15221
|
507
|
-
sglang/srt/models/mllama.py,sha256=
|
515
|
+
sglang/srt/models/mllama.py,sha256=SsK_cEolaeoXh_HkyXsSF2ueYR3sPv1NvnGH2k6Aqx0,38461
|
508
516
|
sglang/srt/models/olmo.py,sha256=FJk8A3T3TF5QcTV6rMP8np94QtvxpMWlgCsv_5VwpVE,12632
|
509
517
|
sglang/srt/models/olmo2.py,sha256=U0ScFzWazOrb_Q90sfXkpVNAsXT-pgZbNgGh80R40VE,14288
|
510
518
|
sglang/srt/models/olmoe.py,sha256=tx5OKWLOr6_pohe2eBcIodCmcuSjtpteHq_tG_QVYCY,15910
|
511
519
|
sglang/srt/models/phi3_small.py,sha256=6p-5EBbwN3FmhoL0VNrNb6VP7HqIhYgVBfQ98L6Rjjo,15469
|
512
520
|
sglang/srt/models/qwen.py,sha256=edS0UYq6AoHZdYUJtQa5wyFNzZMW0JAMmBulH2uheaw,10719
|
513
|
-
sglang/srt/models/qwen2.py,sha256=
|
514
|
-
sglang/srt/models/qwen2_5_vl.py,sha256=
|
521
|
+
sglang/srt/models/qwen2.py,sha256=2C5wJXPsaETMGOojZfQ3v2LmqxtVldxu6upZq7ZTqB0,16142
|
522
|
+
sglang/srt/models/qwen2_5_vl.py,sha256=azSqNvzEQT4ykx8f0X8mfh2LTSu2OhO0PEES0K09pro,24329
|
515
523
|
sglang/srt/models/qwen2_classification.py,sha256=dGrMm4ebd30_lBhHOhaV57ig2iOTx3nqB4GEzsrRIM8,2747
|
516
524
|
sglang/srt/models/qwen2_eagle.py,sha256=Iz0HWL2FgSD3FqoFhfYmbIZeEYkPTJ96lYbkncmHJX4,4644
|
517
525
|
sglang/srt/models/qwen2_moe.py,sha256=9cLOPHBpwdID92Ed1CEjMUxMxcWX83lklunHbyK4To0,18206
|
518
526
|
sglang/srt/models/qwen2_rm.py,sha256=-mQXDEv11p-I1HXgYLTtY6ROem6UYorO958WsDrzsgs,2837
|
519
|
-
sglang/srt/models/qwen2_vl.py,sha256=
|
527
|
+
sglang/srt/models/qwen2_vl.py,sha256=HMA6bww6bCYp7hTPUqSOigSCQRbhonKjTS6lxakclAM,22092
|
520
528
|
sglang/srt/models/registry.py,sha256=inKh9iwOp3LFYm3nqujg-OtABClOP-ifc1stA9cZegA,3434
|
521
529
|
sglang/srt/models/stablelm.py,sha256=w93fNXpDwQbuKi4tdeo0bsXFZrMZVY4_pgNL0E5RErQ,12242
|
522
530
|
sglang/srt/models/torch_native_llama.py,sha256=5tfFSMAXB3ScToqTALtCXa8Oo-qPCJh-KQCNB6QOlNA,19293
|
523
531
|
sglang/srt/models/xverse.py,sha256=I7ivNsk6NRqPxlMUmdclpzDCvhAnWbv_GOj01MKHJrQ,13996
|
524
532
|
sglang/srt/models/xverse_moe.py,sha256=xLwn5pRwQrvj7zMmwl3o49m7xILb2ACRdWvm9hY8LDc,16743
|
525
533
|
sglang/srt/models/yivl.py,sha256=oToK7-u5IGO7xwpJIQ7VtudlK6-zPqJX4bt6_wv0SH8,4850
|
526
|
-
sglang/srt/openai_api/adapter.py,sha256=
|
527
|
-
sglang/srt/openai_api/protocol.py,sha256=
|
534
|
+
sglang/srt/openai_api/adapter.py,sha256=DRHA38G0T9EV2npsKZPBBH4RGJocjZtIov3U5d5VDX0,71919
|
535
|
+
sglang/srt/openai_api/protocol.py,sha256=Y8PFFhLbzhpoERM6-WsTkm-ZuGcE-3tfenh9e-AC1vc,13374
|
536
|
+
sglang/srt/platforms/interface.py,sha256=hym3iooBB4C8if5hDZezgVN6h4NIOu7sg2ZUBIV6XmM,11246
|
528
537
|
sglang/srt/sampling/custom_logit_processor.py,sha256=tDvoLgLqn-sy1qcY6vSrpbnHCeqbdk0uhMOO-uy4p4E,1099
|
529
538
|
sglang/srt/sampling/sampling_batch_info.py,sha256=wrGGU27mWOi_yCBBCOvTQLBdyTjfkPuj7Hsk0zOFyH8,11989
|
530
539
|
sglang/srt/sampling/sampling_params.py,sha256=nXm44Inn91YtrMpAm5mDb6-97owRy-Bh6lZ0BIpw73I,5919
|
@@ -534,15 +543,15 @@ sglang/srt/sampling/penaltylib/min_new_tokens.py,sha256=rdU_D7RoIcrQPhysNQEzmr4T
|
|
534
543
|
sglang/srt/sampling/penaltylib/orchestrator.py,sha256=XM-Lm6u7gYPtMZrTIc0FR4QxNZxBH5s_Cj82umyCzYk,5721
|
535
544
|
sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=NRh10AJrrQlGJ6S-enGdRefrTrWpyqrSm-aNnyqQNQQ,2119
|
536
545
|
sglang/srt/speculative/build_eagle_tree.py,sha256=SFQ3eHbhfNxOdxgqDP5wSV_ZlIVqLw7VivycNZ963N0,11690
|
537
|
-
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=
|
538
|
-
sglang/srt/speculative/eagle_utils.py,sha256=
|
539
|
-
sglang/srt/speculative/eagle_worker.py,sha256=
|
546
|
+
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=Bcsp4g0VvBmsrclkgKq512skfw3hkO2zkHX_91pBaAI,9252
|
547
|
+
sglang/srt/speculative/eagle_utils.py,sha256=0kxQ69XNKO52qgKz-afO1aNF5Tbf5g1HHB7GMuUROG8,29074
|
548
|
+
sglang/srt/speculative/eagle_worker.py,sha256=TysB0F6tFjblIjqoD2nlKQPNBMszDilsII7-mFWFjmo,26999
|
540
549
|
sglang/srt/speculative/spec_info.py,sha256=rhaKG0TzyF9XZYHEWp1jccwTBohSNsUDvxHFtAoOl18,709
|
541
550
|
sglang/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
542
551
|
sglang/test/few_shot_gsm8k.py,sha256=7VLbWl4nCQs1wjtW4q-46jf9jUCycSs5Iw8v7sUSzBw,4284
|
543
552
|
sglang/test/few_shot_gsm8k_engine.py,sha256=QQbrwOX6-cJDD3RZC_e7zPnt6aSo8JdF8X_lRHSjdDM,3886
|
544
553
|
sglang/test/run_eval.py,sha256=9yO0hXZOcn4abEOs96T-XPguDEklK16Ltco0pGF3zCg,4020
|
545
|
-
sglang/test/runners.py,sha256=
|
554
|
+
sglang/test/runners.py,sha256=zl_7wdwUbuCSZoA-f94VhnOI36VX_DwCt3cAEzIjm9s,30484
|
546
555
|
sglang/test/send_one.py,sha256=6FhbJ3c8RpXxvFTELRXaF97GpT7zXXsCDYZh1DqG22E,2550
|
547
556
|
sglang/test/simple_eval_common.py,sha256=joqrGysuLnJFtzDRIgFkMsRyKUSyjVPFWp0_PHAL3Ik,12378
|
548
557
|
sglang/test/simple_eval_gpqa.py,sha256=8Xt9Bw05c7SZTYrCZgB68OZUqUbLo69ywiyx0bTvSUk,3220
|
@@ -553,15 +562,15 @@ sglang/test/simple_eval_mmlu.py,sha256=FkwamjGMjueTixymkedF-YiPloSLiy4ftILFUrKZ9
|
|
553
562
|
sglang/test/test_activation.py,sha256=GeTIJHxlLQfW3kM-X1FGa8Sa3dSGKHEXl5wEy-hfGis,1489
|
554
563
|
sglang/test/test_block_fp8.py,sha256=IqdQKt23annq_QR1gwVX0vzdMyWTEBLRhmPiLMemKI8,14458
|
555
564
|
sglang/test/test_block_fp8_ep.py,sha256=N1rvqbPErBaFFpeAw8TLYXGNZOoG7cfIBP2p5XbSyMo,10806
|
556
|
-
sglang/test/test_custom_ops.py,sha256=
|
565
|
+
sglang/test/test_custom_ops.py,sha256=4X3-odkJntwNtBAuKtCbYHu6peIP6LaI_VwLw7kmDx8,5550
|
557
566
|
sglang/test/test_dynamic_grad_mode.py,sha256=L76yUCuk_ymNpXD2CmO8r2GiGjIvD_gtTsuFDs2NolI,1638
|
558
567
|
sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg,3764
|
559
568
|
sglang/test/test_programs.py,sha256=VZ3vXtUDBnXz0M7gFdDH8hXg9Wa0j_qI8CVqjEgRN_E,18877
|
560
|
-
sglang/test/test_utils.py,sha256=
|
569
|
+
sglang/test/test_utils.py,sha256=jUkIDxJ7I8hCPk0XF7F_IWJkOtn6O7eXJG5pI0cduwo,30463
|
561
570
|
sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
562
571
|
sglang/test/attention/test_flashattn_backend.py,sha256=OxS1KsPs19nwZcDtdURj7_liT1cIfEXb6W4FH9KMaaE,10808
|
563
|
-
sglang-0.4.4.
|
564
|
-
sglang-0.4.4.
|
565
|
-
sglang-0.4.4.
|
566
|
-
sglang-0.4.4.
|
567
|
-
sglang-0.4.4.
|
572
|
+
sglang-0.4.4.post4.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
|
573
|
+
sglang-0.4.4.post4.dist-info/METADATA,sha256=gHc8ZNyeSnw_PrtGOQ7SAVJwCqonaOP4k0vADEpMqv4,25067
|
574
|
+
sglang-0.4.4.post4.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
|
575
|
+
sglang-0.4.4.post4.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
|
576
|
+
sglang-0.4.4.post4.dist-info/RECORD,,
|
File without changes
|
File without changes
|
File without changes
|