sglang 0.4.3.post3__py3-none-any.whl → 0.4.3.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 +1 -1
- sglang/srt/layers/attention/flashinfer_backend.py +94 -48
- sglang/srt/layers/attention/triton_backend.py +4 -2
- sglang/srt/managers/io_struct.py +1 -0
- sglang/srt/managers/scheduler.py +144 -127
- sglang/srt/managers/tokenizer_manager.py +1 -0
- sglang/srt/mem_cache/memory_pool.py +34 -29
- sglang/srt/metrics/collector.py +8 -0
- sglang/srt/model_executor/cuda_graph_runner.py +1 -7
- sglang/srt/model_executor/model_runner.py +97 -78
- sglang/srt/server_args.py +3 -12
- sglang/srt/speculative/build_eagle_tree.py +6 -1
- sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +1 -11
- sglang/srt/speculative/eagle_utils.py +2 -1
- sglang/srt/speculative/eagle_worker.py +67 -32
- sglang/version.py +1 -1
- {sglang-0.4.3.post3.dist-info → sglang-0.4.3.post4.dist-info}/METADATA +2 -1
- {sglang-0.4.3.post3.dist-info → sglang-0.4.3.post4.dist-info}/RECORD +21 -21
- {sglang-0.4.3.post3.dist-info → sglang-0.4.3.post4.dist-info}/LICENSE +0 -0
- {sglang-0.4.3.post3.dist-info → sglang-0.4.3.post4.dist-info}/WHEEL +0 -0
- {sglang-0.4.3.post3.dist-info → sglang-0.4.3.post4.dist-info}/top_level.txt +0 -0
@@ -3,13 +3,13 @@ sglang/api.py,sha256=rYa2qKE88_RJQwYVvjuJzEZECf75ujchZVqi0q48tqc,6890
|
|
3
3
|
sglang/bench_offline_throughput.py,sha256=OQb-AjL4UNymmir02ht43uzgaNsnO_I11nXSowKMqBI,13841
|
4
4
|
sglang/bench_one_batch.py,sha256=mVgmg1LP_Y67HlH4M2V7QvuO1aFjCE0n3gDRXW_w_NQ,17935
|
5
5
|
sglang/bench_one_batch_server.py,sha256=iu73SsvYwnuRktYZDz1P6psMiRx8MbEbF5sbsYJdzYg,5962
|
6
|
-
sglang/bench_serving.py,sha256=
|
6
|
+
sglang/bench_serving.py,sha256=DGpC7L7fH3F8h_yOvkhMWwSxPQnNaYInE9gSv13Xeb8,55327
|
7
7
|
sglang/check_env.py,sha256=lDVA3ybt1wOE33HIMpkkU7zGRgLWez1_ifRRJ8qxbtw,8445
|
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=hhoF2AILUCu0p7La-7pYAF2C6FeeaXGvErj3SqJTYxg,15404
|
12
|
-
sglang/version.py,sha256=
|
12
|
+
sglang/version.py,sha256=ZlAh3ELJql74nlCek86guWjIlTdZddRBNjrR9kE8_Dk,28
|
13
13
|
sglang/lang/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
14
14
|
sglang/lang/chat_template.py,sha256=0tZX67LgtYGrWopnSuTeqWVdxaw2deJOFWOBJpd6htU,17547
|
15
15
|
sglang/lang/choices.py,sha256=-W1DVw9N9ZliVpvmWrzIXG4cswAah8eMQrHWzkS3D8o,6234
|
@@ -34,7 +34,7 @@ sglang/srt/mm_utils.py,sha256=1ScBunw_x4W8ebM_AcJ62-1T2mfT8NlMJqdAhkF1lb0,12367
|
|
34
34
|
sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
|
35
35
|
sglang/srt/reasoning_parser.py,sha256=45xsU9RCPfyG4_Zx4y3-JPyNgAtrqwKI4j5R2NT4g1s,5594
|
36
36
|
sglang/srt/server.py,sha256=PrQb9r6L9syWHKlggbbiQYsKtpwSmECqozRbf8qnoV8,874
|
37
|
-
sglang/srt/server_args.py,sha256=
|
37
|
+
sglang/srt/server_args.py,sha256=61p3vmiMcw3I-g_Xfs2OsMRKroxqzJkGdf_zerRN2js,44682
|
38
38
|
sglang/srt/torch_memory_saver_adapter.py,sha256=--FgbrcvJxTcRe856plD9ktqgrHGPTE18eZCJlE50hY,1255
|
39
39
|
sglang/srt/utils.py,sha256=7P7fyymG10E9ImmjUlRDVuJa8zn9C2tfnxULfw1CP30,47673
|
40
40
|
sglang/srt/warmup.py,sha256=FmJiYfjRr3X_eAe7ojQaPoN17LvHpjDmRWRnO-k86AQ,1469
|
@@ -80,10 +80,10 @@ sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSa
|
|
80
80
|
sglang/srt/layers/vocab_parallel_embedding.py,sha256=WFMCGR4EdYloQK5fT1F6BbA7rT5OK76_7pzV81wiJWU,22286
|
81
81
|
sglang/srt/layers/attention/base_attn_backend.py,sha256=j6pDIDuxlWVmc8GzNxcAFi1USRo4gqqI7G65XAQ5u5M,3263
|
82
82
|
sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
|
83
|
-
sglang/srt/layers/attention/flashinfer_backend.py,sha256=
|
83
|
+
sglang/srt/layers/attention/flashinfer_backend.py,sha256=S8t6u0PliXhtUqnDIG8uYU2h1aIhujkJHyqawxWrYPo,45962
|
84
84
|
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=De7VK-2-E5cjuwffOFq0KPjQtDfnJnkFJDdOp-nZbSQ,20003
|
85
85
|
sglang/srt/layers/attention/torch_native_backend.py,sha256=KABmBrMqKa4x08kkQYdIcZUGydvmaVJIUfo3y8jhFHI,9270
|
86
|
-
sglang/srt/layers/attention/triton_backend.py,sha256=
|
86
|
+
sglang/srt/layers/attention/triton_backend.py,sha256=3bt0Cs0B3w7QURtq7AdBgTjcGCAj_Ojod3kjWIdk58k,21542
|
87
87
|
sglang/srt/layers/attention/utils.py,sha256=gwZoJDyJ9OIPMHWl6r3qkuyVp4Sji9juX7Pwvh9PNxI,1131
|
88
88
|
sglang/srt/layers/attention/vision.py,sha256=rDKvuA14zC8JQaKJFKZSf0hBbpf3F-ELTzcFk0y3l7o,12939
|
89
89
|
sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=fmUCxXnnWt9FlomBrveedq5WB-Gzw87wPncDCoHVA2U,17926
|
@@ -365,12 +365,12 @@ sglang/srt/managers/configure_logging.py,sha256=fOJaXAQ1n9m-8KPJndpsKvS885i69SMa
|
|
365
365
|
sglang/srt/managers/data_parallel_controller.py,sha256=K6jwHn_UhsC7o_lZT5FQm9oKBbsYhlHKiDSutFlm3jA,9539
|
366
366
|
sglang/srt/managers/detokenizer_manager.py,sha256=HTfpJWMF1EImhKOnLJ96xPmYXm71xzaisLMfxg3zpgs,10111
|
367
367
|
sglang/srt/managers/image_processor.py,sha256=NQnhbV8AYr9H6wTKulxcBd6cauBlkySNSDQzk6nKs74,23906
|
368
|
-
sglang/srt/managers/io_struct.py,sha256=
|
368
|
+
sglang/srt/managers/io_struct.py,sha256=ltod5DPNQBJalrz420mcSZReSV1EmYQF9Edh_FiDMf4,22047
|
369
369
|
sglang/srt/managers/schedule_batch.py,sha256=oYrAtlDUjiekQ8xTSnR15i19SJcNyfbiy61PBJpbMi0,52241
|
370
370
|
sglang/srt/managers/schedule_policy.py,sha256=CTN9bggZ8_EiJJYJwEE6rSlfSrrVjka-hadXcQn10HM,18321
|
371
|
-
sglang/srt/managers/scheduler.py,sha256=
|
371
|
+
sglang/srt/managers/scheduler.py,sha256=Kd80MjJEKxSfsETlacUtOZmZtF5iUI-ED3d8G8QcvHo,92144
|
372
372
|
sglang/srt/managers/session_controller.py,sha256=YOv8cFcuVmdCE4OfQJ6aA5AosHRwtZF9WdUUfUXEp0I,5753
|
373
|
-
sglang/srt/managers/tokenizer_manager.py,sha256=
|
373
|
+
sglang/srt/managers/tokenizer_manager.py,sha256=Wg_W42A04WBVTDUpRYpvIW2-RK-AiMwZ9-ytCN_2PjM,44750
|
374
374
|
sglang/srt/managers/tp_worker.py,sha256=o9MY1a8x81nI3W0m64YvOXcAA4sCBvPPmnNU67vBXGs,8710
|
375
375
|
sglang/srt/managers/tp_worker_overlap_thread.py,sha256=bFt8L6H1Tz3QWYdVx9Hl9vUbxY8xmAiaMEMFzCwd7Nk,9022
|
376
376
|
sglang/srt/managers/utils.py,sha256=5i75uLlQOF_5CaT02CrWtwozMTtwTg2_nLP8Dtr-JZQ,1536
|
@@ -378,13 +378,13 @@ sglang/srt/mem_cache/base_prefix_cache.py,sha256=qxgpSHm3qtMdab4U35Mr2BE9TQNjElr
|
|
378
378
|
sglang/srt/mem_cache/chunk_cache.py,sha256=gsXvfCkMRLHIvInFopoSGoDxCDd_um7VDPZaWpUF0v4,2696
|
379
379
|
sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
|
380
380
|
sglang/srt/mem_cache/hiradix_cache.py,sha256=x8cX2d88QjbEjqyOF3vT67lyFSdLem8AYAxej9NF5Jw,14256
|
381
|
-
sglang/srt/mem_cache/memory_pool.py,sha256=
|
381
|
+
sglang/srt/mem_cache/memory_pool.py,sha256=ovqpob1luqfzg3qckANwUnEcpLTG-DJcULIZQyeFlVQ,20687
|
382
382
|
sglang/srt/mem_cache/radix_cache.py,sha256=LbdorXpJ42mmNv9y_Mll0v_c5hgy0LraDCn8vK78QnA,12161
|
383
|
-
sglang/srt/metrics/collector.py,sha256=
|
383
|
+
sglang/srt/metrics/collector.py,sha256=0X40ZZ18182sx2t0eqeqoK7gspH36L343zNvSkgBvd0,9293
|
384
384
|
sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
|
385
|
-
sglang/srt/model_executor/cuda_graph_runner.py,sha256=
|
385
|
+
sglang/srt/model_executor/cuda_graph_runner.py,sha256=l4coS7de4zwYbxuBokZ3MyLZUx9ygGW1BqqqMCJGNtQ,20010
|
386
386
|
sglang/srt/model_executor/forward_batch_info.py,sha256=8x3y5rCMotL8iSoSG3YMd77bI6mZOoisbHIbJcBZT04,16809
|
387
|
-
sglang/srt/model_executor/model_runner.py,sha256=
|
387
|
+
sglang/srt/model_executor/model_runner.py,sha256=b1NEnYaKbg_w6fu2cjdX_YdMjkpwRgBgfMf86T1JJ7w,41727
|
388
388
|
sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
|
389
389
|
sglang/srt/model_loader/loader.py,sha256=zDmZ7NZ9Z44EDR2H8d3bKadZD4Ey7Jmju8wBnWhM1FI,46805
|
390
390
|
sglang/srt/model_loader/utils.py,sha256=0NaMR67fESFopaklmsleiL27XH1QUrjZW246MUu1EJ0,1369
|
@@ -447,10 +447,10 @@ sglang/srt/sampling/penaltylib/frequency_penalty.py,sha256=wdWLmhYnJkuS5qjFGbSLs
|
|
447
447
|
sglang/srt/sampling/penaltylib/min_new_tokens.py,sha256=rdU_D7RoIcrQPhysNQEzmr4TO2OoEi___p-i3QdwkgU,3331
|
448
448
|
sglang/srt/sampling/penaltylib/orchestrator.py,sha256=XM-Lm6u7gYPtMZrTIc0FR4QxNZxBH5s_Cj82umyCzYk,5721
|
449
449
|
sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=ZU18IAOSd1qNcZdsP47TS-gjM-jVq9s8YuQWhUZ7xZI,2205
|
450
|
-
sglang/srt/speculative/build_eagle_tree.py,sha256=
|
451
|
-
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=
|
452
|
-
sglang/srt/speculative/eagle_utils.py,sha256=
|
453
|
-
sglang/srt/speculative/eagle_worker.py,sha256=
|
450
|
+
sglang/srt/speculative/build_eagle_tree.py,sha256=8_uUpkQAE0qcn5mA6NPjfl81EMuNxg4fZq628wjEZNU,20805
|
451
|
+
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=zsF4AcsykaFjzZ5SKdvUQyEB1GVXTvR1_kI_oaQafBo,7704
|
452
|
+
sglang/srt/speculative/eagle_utils.py,sha256=DhpNNNN56jw0ucXSBYq9_IzDuJmEKgXS6lbLczaghNs,27196
|
453
|
+
sglang/srt/speculative/eagle_worker.py,sha256=05bUiqQdO6B9dwfSW1BOEtez6HX24OHwJfm6bFPJa7I,18807
|
454
454
|
sglang/srt/speculative/spec_info.py,sha256=hJR0b3gZ0QA0KZLq6TfqSDJkmpV6mThyle3sHYI4o4M,522
|
455
455
|
sglang/test/few_shot_gsm8k.py,sha256=7VLbWl4nCQs1wjtW4q-46jf9jUCycSs5Iw8v7sUSzBw,4284
|
456
456
|
sglang/test/few_shot_gsm8k_engine.py,sha256=QQbrwOX6-cJDD3RZC_e7zPnt6aSo8JdF8X_lRHSjdDM,3886
|
@@ -469,8 +469,8 @@ sglang/test/test_block_fp8_ep.py,sha256=hkuQjmCv3y_hWZj21cT9EaB6KSfT3JSzYPRQNFaL
|
|
469
469
|
sglang/test/test_layernorm.py,sha256=IacByD5d-stXjzBz8Ypamc7povlcedpKPbb_4JLgo3c,3720
|
470
470
|
sglang/test/test_programs.py,sha256=VZ3vXtUDBnXz0M7gFdDH8hXg9Wa0j_qI8CVqjEgRN_E,18877
|
471
471
|
sglang/test/test_utils.py,sha256=jVkIdnzkQ5ZGynWkfFMd6GLonJwq_2a6iZAvPLUFlZo,28549
|
472
|
-
sglang-0.4.3.
|
473
|
-
sglang-0.4.3.
|
474
|
-
sglang-0.4.3.
|
475
|
-
sglang-0.4.3.
|
476
|
-
sglang-0.4.3.
|
472
|
+
sglang-0.4.3.post4.dist-info/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
|
473
|
+
sglang-0.4.3.post4.dist-info/METADATA,sha256=I3MCfxnWrPkIIQdq1rdqEqm0_1QkyR_QmVZ-jcIbHZU,24409
|
474
|
+
sglang-0.4.3.post4.dist-info/WHEEL,sha256=jB7zZ3N9hIM9adW7qlTAyycLYW9npaWKLRzaoVcLKcM,91
|
475
|
+
sglang-0.4.3.post4.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
|
476
|
+
sglang-0.4.3.post4.dist-info/RECORD,,
|
File without changes
|
File without changes
|
File without changes
|