sglang 0.3.4.post2__py3-none-any.whl → 0.3.5__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.
Files changed (78) hide show
  1. sglang/api.py +1 -1
  2. sglang/bench_latency.py +3 -3
  3. sglang/bench_server_latency.py +2 -3
  4. sglang/bench_serving.py +92 -0
  5. sglang/global_config.py +9 -3
  6. sglang/lang/chat_template.py +50 -25
  7. sglang/lang/interpreter.py +9 -1
  8. sglang/lang/ir.py +11 -2
  9. sglang/launch_server.py +1 -1
  10. sglang/srt/configs/model_config.py +51 -13
  11. sglang/srt/constrained/__init__.py +18 -0
  12. sglang/srt/constrained/bnf_cache.py +61 -0
  13. sglang/srt/constrained/grammar.py +190 -0
  14. sglang/srt/hf_transformers_utils.py +6 -5
  15. sglang/srt/layers/attention/triton_ops/decode_attention.py +110 -30
  16. sglang/srt/layers/attention/triton_ops/prefill_attention.py +1 -1
  17. sglang/srt/layers/fused_moe/fused_moe.py +4 -3
  18. sglang/srt/layers/fused_moe/layer.py +28 -0
  19. sglang/srt/layers/quantization/base_config.py +16 -1
  20. sglang/srt/layers/vocab_parallel_embedding.py +486 -0
  21. sglang/srt/managers/data_parallel_controller.py +7 -6
  22. sglang/srt/managers/detokenizer_manager.py +9 -11
  23. sglang/srt/managers/image_processor.py +4 -3
  24. sglang/srt/managers/io_struct.py +70 -78
  25. sglang/srt/managers/schedule_batch.py +33 -49
  26. sglang/srt/managers/schedule_policy.py +24 -13
  27. sglang/srt/managers/scheduler.py +137 -80
  28. sglang/srt/managers/tokenizer_manager.py +224 -336
  29. sglang/srt/managers/tp_worker.py +5 -5
  30. sglang/srt/mem_cache/flush_cache.py +1 -1
  31. sglang/srt/model_executor/cuda_graph_runner.py +7 -4
  32. sglang/srt/model_executor/model_runner.py +8 -17
  33. sglang/srt/models/baichuan.py +4 -4
  34. sglang/srt/models/chatglm.py +4 -4
  35. sglang/srt/models/commandr.py +1 -1
  36. sglang/srt/models/dbrx.py +5 -5
  37. sglang/srt/models/deepseek.py +4 -4
  38. sglang/srt/models/deepseek_v2.py +4 -4
  39. sglang/srt/models/exaone.py +4 -4
  40. sglang/srt/models/gemma.py +1 -1
  41. sglang/srt/models/gemma2.py +1 -1
  42. sglang/srt/models/gpt2.py +287 -0
  43. sglang/srt/models/gpt_bigcode.py +1 -1
  44. sglang/srt/models/grok.py +4 -4
  45. sglang/srt/models/internlm2.py +4 -4
  46. sglang/srt/models/llama.py +15 -7
  47. sglang/srt/models/llama_embedding.py +2 -10
  48. sglang/srt/models/llama_reward.py +5 -0
  49. sglang/srt/models/minicpm.py +4 -4
  50. sglang/srt/models/minicpm3.py +4 -4
  51. sglang/srt/models/mixtral.py +7 -5
  52. sglang/srt/models/mixtral_quant.py +4 -4
  53. sglang/srt/models/mllama.py +5 -5
  54. sglang/srt/models/olmo.py +4 -4
  55. sglang/srt/models/olmoe.py +4 -4
  56. sglang/srt/models/qwen.py +4 -4
  57. sglang/srt/models/qwen2.py +4 -4
  58. sglang/srt/models/qwen2_moe.py +4 -4
  59. sglang/srt/models/qwen2_vl.py +4 -8
  60. sglang/srt/models/stablelm.py +4 -4
  61. sglang/srt/models/torch_native_llama.py +4 -4
  62. sglang/srt/models/xverse.py +4 -4
  63. sglang/srt/models/xverse_moe.py +4 -4
  64. sglang/srt/openai_api/adapter.py +52 -66
  65. sglang/srt/sampling/sampling_batch_info.py +7 -13
  66. sglang/srt/server.py +31 -35
  67. sglang/srt/server_args.py +34 -5
  68. sglang/srt/utils.py +40 -56
  69. sglang/test/runners.py +2 -1
  70. sglang/test/test_utils.py +73 -25
  71. sglang/utils.py +62 -1
  72. sglang/version.py +1 -1
  73. sglang-0.3.5.dist-info/METADATA +344 -0
  74. {sglang-0.3.4.post2.dist-info → sglang-0.3.5.dist-info}/RECORD +77 -73
  75. {sglang-0.3.4.post2.dist-info → sglang-0.3.5.dist-info}/WHEEL +1 -1
  76. sglang-0.3.4.post2.dist-info/METADATA +0 -899
  77. {sglang-0.3.4.post2.dist-info → sglang-0.3.5.dist-info}/LICENSE +0 -0
  78. {sglang-0.3.4.post2.dist-info → sglang-0.3.5.dist-info}/top_level.txt +0 -0
@@ -1,20 +1,20 @@
1
1
  sglang/__init__.py,sha256=b_pqO9bR2fjK9En_tigfzKTiQzE8b_hUizY0DAKVk1M,1616
2
- sglang/api.py,sha256=5x591S4rLbmNPs75qPwGKVu1sonVGDyjPAJlHTyWw50,6956
3
- sglang/bench_latency.py,sha256=L1zh6Gobr6A2gTsBMjlDgaH36RATwQ8_zbEPRogsC9s,18288
4
- sglang/bench_server_latency.py,sha256=2AMPwU2_85q-Btz9UdZC-TnZJPgXcNkydvFYWn2CJlU,5892
5
- sglang/bench_serving.py,sha256=jcxNP7reIJPh3x1hG5TCM6wMlDXshjyMJUUjL2O7kzs,40060
2
+ sglang/api.py,sha256=3I9YUJNOeCqwKymZec2JR_agjTyKIx4XoT6IGdZ4_Cs,6953
3
+ sglang/bench_latency.py,sha256=SSqZjcCNO88ExpT94qBZ5CmuA5o0T8wMTBnxLsNMqik,18259
4
+ sglang/bench_server_latency.py,sha256=N1MODIzcMk74yOWmY19d36aih3ewtHOemLxoieKtdhw,5866
5
+ sglang/bench_serving.py,sha256=0RR0RsrQqLWqcIPENfrS97F9HJiVXIZvGOWy4R2GvDA,43680
6
6
  sglang/check_env.py,sha256=rGRABCgt-0SfUrow4px28b2P59aMn8eVTnN5eZc_a8s,5397
7
- sglang/global_config.py,sha256=1r_W9rrBxGCCc2eqESRduOMMNq46e54xLgFLifHuQm0,1014
8
- sglang/launch_server.py,sha256=UnjNjYuZ8TtvmRtgYEsFImkbvCwvn_tQjk0V7cHy67E,450
7
+ sglang/global_config.py,sha256=fnT0U9vlHdGaQFKN9tYTnUF4-eVW4HYQURd5zvPtrg0,1286
8
+ sglang/launch_server.py,sha256=_XIqBcXArYtHTqilOFkYWKZBYXGCMHAxbYOST08LGj0,415
9
9
  sglang/launch_server_llavavid.py,sha256=tGc17S1vUfLwbi1GB26oOdXxTWr7gjlqpTrPnrMRNO8,1007
10
- sglang/utils.py,sha256=NA_4xUrTI7KICQ3PEACfNWKE3nxSA5QvQZJNd4TQrDc,9395
11
- sglang/version.py,sha256=XbWUUFDK_Ls5pw5MPNOhsEKwI18aKVrvco5r4BTbEgU,28
10
+ sglang/utils.py,sha256=73tkeT4gDzmVkWO4nVXQHS9XlzH7CSL-I_uRpEDsCPg,11546
11
+ sglang/version.py,sha256=ThnCuF3X7rsQSd5PAea_jfYA70ZmhLvkFcLBxBPwZnY,22
12
12
  sglang/lang/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
13
- sglang/lang/chat_template.py,sha256=pmSCliE2vLCwx6m588Nv6QybNPc_o_CzSDS_YFmCmEo,14051
13
+ sglang/lang/chat_template.py,sha256=jprS3-In2FTUoedKwZg-HYvDwU8RTIYntOlf2zoN2sU,14814
14
14
  sglang/lang/choices.py,sha256=-W1DVw9N9ZliVpvmWrzIXG4cswAah8eMQrHWzkS3D8o,6234
15
15
  sglang/lang/compiler.py,sha256=o1C6G3TzhjSlsH-doTPy5oiVehr57dxNTa5oZw5TTAI,7639
16
- sglang/lang/interpreter.py,sha256=zakc6IkzATaMqVDWKWvqDRrqnRykxFawajA7aUHUDbI,30640
17
- sglang/lang/ir.py,sha256=F_9ac10OjktxR7KhOV07wiJXV20s79cRfh9d4koExJc,18262
16
+ sglang/lang/interpreter.py,sha256=SBjejhLhTKzNM0HbjtTg5r17WPJ64WFSk6lcM_SCWKs,30717
17
+ sglang/lang/ir.py,sha256=zpzzAO1YVldhE95Vwz5hU_TQltu-xt8A6rfFr0PuIDA,18410
18
18
  sglang/lang/tracer.py,sha256=borJmlSJOhg1RUndGRnilnR60eEZz2Y9aU7BpftsOxU,8287
19
19
  sglang/lang/backend/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
20
20
  sglang/lang/backend/anthropic.py,sha256=EXRX7xJgA5KZszX7toSLVnKzFQ5EO0Loj-YjHFtxSxg,2081
@@ -24,18 +24,20 @@ sglang/lang/backend/openai.py,sha256=qM7eVH_kMxnDd2rpxOH0v76KxtOJFlAwgLgWIKvFGCI
24
24
  sglang/lang/backend/runtime_endpoint.py,sha256=iVb7SlrpJ1ic92QG5kQUphZUb2EaVWY43dkmAO5pju4,10514
25
25
  sglang/lang/backend/vertexai.py,sha256=O-iBLD-y3vq80UxnrAoJri7bxpgd-_eakZ88Cf8bEGA,4855
26
26
  sglang/srt/conversation.py,sha256=erz6wEXMcSmBlskuUhX2c-MT0EMyqyFpTem9PgastEE,21107
27
- sglang/srt/hf_transformers_utils.py,sha256=3G41c2HbEsI-aN6VtbSo4MyqVhM5ZtTTiRM9NQ6jMzQ,6597
27
+ sglang/srt/hf_transformers_utils.py,sha256=QbYVTnz0UdaXESPMAaq1OMzzznn95J_l08eXJuB68aU,6618
28
28
  sglang/srt/mm_utils.py,sha256=zox644S3IHUWmADdK4MnIbdTS2DWHOy0_Dq0gCU38QQ,12273
29
- sglang/srt/server.py,sha256=uagMRPk-9n2mpOXbLhjiBSK3nGqGwh3ZyzKu8ZnsDUg,27204
30
- sglang/srt/server_args.py,sha256=AQHVnYJs62FBCFAvq4KYZtdvqGi0iOr41aFuHpb2qQw,27813
31
- sglang/srt/utils.py,sha256=t7GqQXzgGqJ_9v0_gEgr8v4_6zf86OHXthEvVe13Rsw,23710
29
+ sglang/srt/server.py,sha256=4yKD85OlhhkneF7VOzWZMro0P8n1xdKgnZfCWdjrXao,27502
30
+ sglang/srt/server_args.py,sha256=AfbBXcrC_XpTWOoZcace0iRksKwyh8-NS1E7RMTWM5A,28912
31
+ sglang/srt/utils.py,sha256=zdoZlo0_R18mAWFc4tYnkxVb7qhqcCTKovaEn2dAHLw,23121
32
32
  sglang/srt/configs/__init__.py,sha256=_usVIXHQjft4PAJ1Y-yGQOn2QNOv501GYMlQwpGXbns,208
33
33
  sglang/srt/configs/exaone.py,sha256=Duxd4yQoKy8GWEzZD_kCY_OzmN_67CTJL_Kgn0eXk3g,10731
34
- sglang/srt/configs/model_config.py,sha256=gSRANcpoB9o5je9a_atYokIo9CsRBJhAtg8FmZbLmSo,8026
34
+ sglang/srt/configs/model_config.py,sha256=bxG-vD8ZmXbypKW6Hvz8AS7rcwjTxt0TzG7p59m3t2E,9387
35
35
  sglang/srt/configs/qwen2vl.py,sha256=AYHuFgJ0bwhWYkD7S6fvP7yJejJnuhy4xp5Q2W-O6ps,4424
36
- sglang/srt/constrained/__init__.py,sha256=ze8awDPvwAzdeMwzJ-25kXOQ4nVWoaP55jBDt5UOS_4,2070
36
+ sglang/srt/constrained/__init__.py,sha256=VXEY9K8HrEBv6QHe3X7J5ingiDugSF9_cpEbEcpBId4,2466
37
37
  sglang/srt/constrained/base_tool_cache.py,sha256=5sazBMHHDpHMoqOjuY6itCxwTmIFCflIWEDXMtmrPVs,2006
38
+ sglang/srt/constrained/bnf_cache.py,sha256=c8msJ57Gj7aMy1ccTRERLgVuovEeDJx-wVPOhYF0w9k,2057
38
39
  sglang/srt/constrained/fsm_cache.py,sha256=CWwtOHTU3sHmw71OhWxl05YgU7cNNVWohlUt71rG230,3536
40
+ sglang/srt/constrained/grammar.py,sha256=kvfyP2E53xo8jVWVZ_qHlJn0U4Qi2WaNi2yMZPKgI_0,6952
39
41
  sglang/srt/constrained/jump_forward.py,sha256=o-CzJu3DEs0eFKlLzsQVYMSo4vBKpffs25sXLOJd6jc,6997
40
42
  sglang/srt/layers/activation.py,sha256=7VEkCrx2dvl629Lz0fkJcJfVoZA-ykEdkpTzKEc_drQ,5225
41
43
  sglang/srt/layers/layernorm.py,sha256=HCj8Y_X6MNNdtQU2sWKgyjIqVERxl9dqrmjbBbyJjpE,3796
@@ -46,79 +48,81 @@ sglang/srt/layers/radix_attention.py,sha256=i07VRXPDHj-zJ1TSrXEqCxumQwYSHwAvc8Do
46
48
  sglang/srt/layers/rotary_embedding.py,sha256=gfRKBB8FmsQKiDH0Crh_KRIGRUuvEgazH1p_n9D_m7E,3889
47
49
  sglang/srt/layers/sampler.py,sha256=3zfth1Kz24X4sUq7Z_cjZwHgPVivI-rgPtIeUbsiiWU,4589
48
50
  sglang/srt/layers/torchao_utils.py,sha256=1nzZkSzbF4qCAMeBKAeeDpMl_mK8imiY2RL3xFEgvAw,3340
51
+ sglang/srt/layers/vocab_parallel_embedding.py,sha256=8Tx0WUNibDoNkGruGzRIkvp6t7D54e-nchdezeQ5Nzk,22302
49
52
  sglang/srt/layers/attention/__init__.py,sha256=EL1o6Q5vLgViN3pOr2A7F6K9FlNEpMdBypFAVMeq_HA,2445
50
53
  sglang/srt/layers/attention/double_sparsity_backend.py,sha256=BlX7uXteQpnoOnKsdBKh8h20zMVMEiibB5F_PkZSlNI,10706
51
54
  sglang/srt/layers/attention/flashinfer_backend.py,sha256=843CbZsRfzWp5FTusNXXL1o4N3jd0hoCNpsoUR6Qjxk,23306
52
55
  sglang/srt/layers/attention/triton_backend.py,sha256=DKUEzxQE8iBvJPNHmQwP1pyx2wXmSsLqzBhLjJznIUk,6482
53
- sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=XCQTX0kUttT1AG5FRMgfQbiXgvoempYD0UR2r6D_vJg,16711
56
+ sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=7cDNPMMkz7--ebNKUeSaLY_6hBbvr_NqDodYFtW9ahA,18433
54
57
  sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=1pSXfY3EEaM7iRN_uElHnAfsrJMhTFbu9fj8Z0O2PbE,21480
55
58
  sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=nEG7iBh1pAy3WaqPdLZwCJwDgyk5HLQ181kBS2nxbwg,11179
56
- sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=MCZMC2XQBx-THzbygpa-nfqXe_ZZJhxAnqntLkmEhPI,5994
59
+ sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=LnuWqGAba03e25adxS_lFgjTV6nBWsVBUGUvrl-8alQ,5993
57
60
  sglang/srt/layers/fused_moe/__init__.py,sha256=bWCrDdOy2ANEXTb8CHYO63O3Iu3eZnn0PJbgl0z5vvE,75
58
- sglang/srt/layers/fused_moe/fused_moe.py,sha256=1WM2cObWXcFWtqh_utGJFPnrT344rORwuQ9hJDaH2s0,23104
59
- sglang/srt/layers/fused_moe/layer.py,sha256=raFyvPzjYz-Fv8B3IcOxQYKKCWqXis5mXwg1GFE61y4,22243
61
+ sglang/srt/layers/fused_moe/fused_moe.py,sha256=uRmDUleTaJKBbsTfum6RgHifUbgi6yKuB2dw_mIhw3M,23250
62
+ sglang/srt/layers/fused_moe/layer.py,sha256=tbHnUJs3uvdDsl3VnwtyGA31VtFouNTPD7h7fPSCYOc,23613
60
63
  sglang/srt/layers/fused_moe/patch.py,sha256=B9cDtHqHfnWE0QqZAffvUi6cVRKcMBMKDGJWGIaKh3U,3898
61
64
  sglang/srt/layers/quantization/__init__.py,sha256=QilMNqgu3eOFUkEjXLSDa1NvoNdi_CAvC8a1hprOgN8,2979
62
- sglang/srt/layers/quantization/base_config.py,sha256=vlpSPvSrFmUe65ETg4SoPocQ9bVNY6As3QuHdr_3Dr4,4023
65
+ sglang/srt/layers/quantization/base_config.py,sha256=fx-FeA1a4jg7HDoYvIKC5G_wLcfeOOyIJQ6MtCaHpZ4,4664
63
66
  sglang/srt/lora/lora.py,sha256=meRL7oBUx8mxV_isc3Lp0EIsFQWC2PvaN-fE78BmMwg,14970
64
67
  sglang/srt/lora/lora_config.py,sha256=paVB7F7SIuxr_vodvKf8zzAlH2fdVYHhXxcXV62D0Vo,1411
65
68
  sglang/srt/lora/lora_manager.py,sha256=gzBwYXZEPYj56PkGTshTbWRfl_370wb6uTcRhDaLiF8,12801
66
- sglang/srt/managers/data_parallel_controller.py,sha256=1MqXT5t9NHsnrTj-b_SUfFNYgKk-6lUdbqpqS1EbftM,5766
67
- sglang/srt/managers/detokenizer_manager.py,sha256=wgFtHwmx336tUNa-BWBmdf7t3qSj0zHpdHI4pmTiqbY,8163
68
- sglang/srt/managers/image_processor.py,sha256=yhrXZvqPq5iSjycez7pA7YyzDS1VAXvCl5IuQ8uKwfs,13793
69
- sglang/srt/managers/io_struct.py,sha256=bEvbibB2xJLFLUxgjCl5VgOBT1VtaH5VAv3zYlirxKY,12403
70
- sglang/srt/managers/schedule_batch.py,sha256=0kZnX1ciaL9yXxUPRyGsEArDvgHnTrZ2V8X4So7VJws,40509
71
- sglang/srt/managers/schedule_policy.py,sha256=unDmK7Y_Ti0Eiizh3_iEFMsC1KDqGMTqU8MlQgg-6qo,11951
72
- sglang/srt/managers/scheduler.py,sha256=e1lYL3Ky8c4Ei6h3lG-r2SavyZlymFMBEn37eFvjsso,43821
73
- sglang/srt/managers/tokenizer_manager.py,sha256=6KaDIzv3yiyRmLOVzvol7rTgnOINKgynLUrRHEXm7Xo,25980
74
- sglang/srt/managers/tp_worker.py,sha256=LbVXE6GiCnmDr117HjB-lxVbFqLL71YEM9y0Q46nWns,5733
69
+ sglang/srt/managers/data_parallel_controller.py,sha256=_XB6Ianc8TiqwLTW-7DH6gGjVYBeBU_6WjjaDk0snIY,5686
70
+ sglang/srt/managers/detokenizer_manager.py,sha256=pBCcK-wKgPk4Ty-vQFSGovEZEE_yKK1f7YVDW8vDcYw,7962
71
+ sglang/srt/managers/image_processor.py,sha256=Pk_dtXzljTkFt7Acsv1RyDzEqvCvjc7BMngxGhtkpDU,13817
72
+ sglang/srt/managers/io_struct.py,sha256=23-eJQrpMw7OJ0LiDvBVKpI36rdyxJluFlHJ7wXjKqw,12261
73
+ sglang/srt/managers/schedule_batch.py,sha256=LIkxGNZC_PWIX7-BJGLRpzgNIGH-1ZxL9RUZE-dgo70,39653
74
+ sglang/srt/managers/schedule_policy.py,sha256=LH0rh1PiI5LK-dSd3dar8_po6FidiBUuj0Xcp_yNQAA,12295
75
+ sglang/srt/managers/scheduler.py,sha256=p72s46nNnUl5YTKfgwRNmcc8NZbBSGudYuqOP2bZsyc,45524
76
+ sglang/srt/managers/tokenizer_manager.py,sha256=fGVMxJb-UQPokqdlbphWHSVnLyKWAY8JK7fHe6iVa2I,21793
77
+ sglang/srt/managers/tp_worker.py,sha256=S5oim5xrkg1j68hYq6LfC8T533JYmQX9Kabt6U8ZXn4,5726
75
78
  sglang/srt/managers/tp_worker_overlap_thread.py,sha256=j5J4yHyR7w2HgAbN7S__299ADvsoyap5HK63SWMNavQ,7546
76
79
  sglang/srt/mem_cache/base_prefix_cache.py,sha256=qEQwEkG4E5rab2ZoTqcesf5pR_J4nV2jBxIHsBJHtIM,924
77
80
  sglang/srt/mem_cache/chunk_cache.py,sha256=VcCpyrf5FOQ5xoKeOouCI5ZQLkZo_pgY1SPbDDkagGg,2492
78
- sglang/srt/mem_cache/flush_cache.py,sha256=pTLKPRB17U6vl5RFJJvuJ4jCL2SyomgkUBNlkDpGRqo,978
81
+ sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
79
82
  sglang/srt/mem_cache/memory_pool.py,sha256=41fjuj_sD0yfJq-sy-X99cc2djBa6w4dy2y47V0WqNU,10934
80
83
  sglang/srt/mem_cache/radix_cache.py,sha256=DzLCO_gYQ7X_C2NJSEHzzMZhb5HzWjKF9wXJQsnzr8M,10427
81
- sglang/srt/model_executor/cuda_graph_runner.py,sha256=t2KPOtoz1k-LhEs8QOHNFgz-tMslI0TSlyBX74WRuaU,12828
84
+ sglang/srt/model_executor/cuda_graph_runner.py,sha256=zRxXxV54b4SUXk9BQ1zPAS2VXCBRBvT15A64Yf0kBSE,12909
82
85
  sglang/srt/model_executor/forward_batch_info.py,sha256=1GM6A-tqTDD0MEMQx93PC7XahABr0vlv7JBXohaehkc,9272
83
- sglang/srt/model_executor/model_runner.py,sha256=90eHOkVSszaXwBgTowCl4pTrzZyJAIosnb2E6MuUXd8,27038
84
- sglang/srt/models/baichuan.py,sha256=uV20fr7SqlATxoziXRtJyXFnZZoWTUy3qvQNFaEvw8M,15014
85
- sglang/srt/models/chatglm.py,sha256=lIu_pVtIWNJ8EsZfr2KZdr2m4GtDTUNvz6D4zkpLMq4,13144
86
- sglang/srt/models/commandr.py,sha256=WIMwjV3C0pRbVs4Xv9tqnHGreRvWC7zsML2hNuXw4A0,14060
87
- sglang/srt/models/dbrx.py,sha256=_DshXyXr_xVB7wtE28PFcb6KDIkA6gygkXYKqivSCFc,14554
88
- sglang/srt/models/deepseek.py,sha256=W342tVpEpkc_fvO_DTP4fX3EGF-DIFC4QOySdUGzl9w,15837
89
- sglang/srt/models/deepseek_v2.py,sha256=Eo3qufIA_WhzCqUhjcJDYWDWOg4REYJvKRmXefL6QoM,28239
90
- sglang/srt/models/exaone.py,sha256=bIXdAXoWlCdfDdX2q47Br3QOa3jEYiiP2Hdd1T4crnM,12993
91
- sglang/srt/models/gemma.py,sha256=4MVHwc5Jc4CSg3HIdNJEBYk8mhspjuwvc_6Oi8Cd-g8,12202
92
- sglang/srt/models/gemma2.py,sha256=3VL223T_3syBG3fUInbtFaXvIs7dYjtsfX3OfDQc7m4,14777
93
- sglang/srt/models/gpt_bigcode.py,sha256=q9N13Js2v0VheudWssRoSjnptS6TSf7DOmC8zLRGxeo,10049
94
- sglang/srt/models/grok.py,sha256=vc7-E_hemNKaNORxg4rmaQcVYlpoavyaAZUG9B2dgbY,14835
95
- sglang/srt/models/internlm2.py,sha256=-liQB13sgR3GnXJacBSMuEbLa2N4tICx0LsNgu_nNvU,12108
96
- sglang/srt/models/llama.py,sha256=a43Y5mvMDmFcRcPL78vsAElaOvTqPajLPB2_BDwJ7pM,15767
86
+ sglang/srt/model_executor/model_runner.py,sha256=Zs-u9sJREJD-1omhaFjBYfgR_2_7Cj0O5mGgQ0NtR8s,26793
87
+ sglang/srt/models/baichuan.py,sha256=RyvPQvi7wy9VUGvLwG17XttcTp43yRj6c3zNRImBToA,15005
88
+ sglang/srt/models/chatglm.py,sha256=9hCXTqGX8DMvSPSn6wlK0YNNRWGS4UiS4-xjFsO9hYU,13135
89
+ sglang/srt/models/commandr.py,sha256=leoQNn4VRqa9SXos6DcrkHVG6-Xp-kjBn2PUgqc9bs8,14051
90
+ sglang/srt/models/dbrx.py,sha256=IiVIk_rVd0RlvfIJGIThPOPkoYT3U649PrduThiKRzg,14545
91
+ sglang/srt/models/deepseek.py,sha256=DjW2B21isWE6A2C8A3VGZ-G0k1DkhWHO3dZZjcOVG50,15828
92
+ sglang/srt/models/deepseek_v2.py,sha256=z6532MRN1tBltFNteFJfimnaGpyNmK6g_sdNmTzsVmk,28230
93
+ sglang/srt/models/exaone.py,sha256=YMyH4zxyCaCB432vCcom800efPI19_vIQ3OXLkLiXxk,12984
94
+ sglang/srt/models/gemma.py,sha256=D_zjG312BeOPeplGzo5Z8tSMH9xL7wZ4KIgczZ9yJ0E,12193
95
+ sglang/srt/models/gemma2.py,sha256=iE56CYzPn-QCis4kcU7Yi0jvJ04KeU2deuZH2DaS2lM,14768
96
+ sglang/srt/models/gpt2.py,sha256=xWqU66KO6rNBnzA6uOBrlLWnwVzLKuC4UWHc5WuEHw8,10151
97
+ sglang/srt/models/gpt_bigcode.py,sha256=f6vvxBFPhV6GIZrOEKjJPu41TyVYw5Knq4h9WDvyEeY,10040
98
+ sglang/srt/models/grok.py,sha256=iSkvt7whYyMndUHBekM4vKHaDXnnmeJMErkklGpz624,14826
99
+ sglang/srt/models/internlm2.py,sha256=HOVOXz3b7eLF2wpG_FEK5PYnYOEpHPGJ0pufvL7HPD0,12099
100
+ sglang/srt/models/llama.py,sha256=X_LKJ02ofDfpgVVFexf_C6g4FikadfMikhuRVAuLN5I,16094
97
101
  sglang/srt/models/llama_classification.py,sha256=WcHYFez7qloTCpXLy1A6-dBGHWp22ebv6yG68jFVBjc,3318
98
- sglang/srt/models/llama_embedding.py,sha256=4j3WNLB-x7XQnJvohdRs7VSSEabbhiE2BRHmnG5IZRU,3453
99
- sglang/srt/models/llama_reward.py,sha256=ag3eVdP38iURj81fTCa-sC2jV_eCkTIjXUQf1I96fCI,5297
102
+ sglang/srt/models/llama_embedding.py,sha256=2ex2jrz31osaAd9V8sJeN0qyxmk-L5NgOBkXL1puGhI,3166
103
+ sglang/srt/models/llama_reward.py,sha256=48J6PmZJRFRv-6mEF6y5fxNKtRRZVQzvJqg3XaWDWa0,5448
100
104
  sglang/srt/models/llava.py,sha256=ny3sK2sgYwrEhawSAc1tZeltcgukphSTdxsqyq-Epkc,24857
101
105
  sglang/srt/models/llavavid.py,sha256=ztS5He-NF4fmfujdoMnKljOG1fNfPvp-6bduT7B6EMU,12137
102
- sglang/srt/models/minicpm.py,sha256=LpUdxKA27z79DSYAPPlfCgI4GEnWCYznhgSQl-QCsTY,13731
103
- sglang/srt/models/minicpm3.py,sha256=-fLZ-RRbR2jLGSsatBWV-qsSNIZCPbS_jasmrOlUdK8,25023
106
+ sglang/srt/models/minicpm.py,sha256=hAzgBImQ1xDeRdaQt5hKcLl1h1T-1QFSerG2MOlLjt8,13722
107
+ sglang/srt/models/minicpm3.py,sha256=O6092exfoq8iHLmyfpVCubyQEzcfp4SmqtZJs7x4A8s,25014
104
108
  sglang/srt/models/mistral.py,sha256=tiYoKjyYVzlQl52QUZ33odD2yCxj9dxcqln474VuZOw,744
105
- sglang/srt/models/mixtral.py,sha256=UUqzpOBXsObirmpJz4xstlG82uu4JfXsh-gWQmiKbW0,13803
106
- sglang/srt/models/mixtral_quant.py,sha256=HPipVG_Gc5Ki0YXg49Rwn2_uvtCCI1IxlA7mVRVFivw,13978
107
- sglang/srt/models/mllama.py,sha256=TW8vyGU_HmyrAlkyHxjy6YAfUKagNDOR6luwKzLtXRc,37898
108
- sglang/srt/models/olmo.py,sha256=lD4VewXK0rVqhttGkOOzaxoqRQgVfV90s8ElStPBBdE,11896
109
- sglang/srt/models/olmoe.py,sha256=3qHnY1DWBhyx9FWGJGb3a8kewcmEdYZOkYZ1JBx1LWs,15251
110
- sglang/srt/models/qwen.py,sha256=mjGqo3NkTYfJ2qqztFw8mjKggPT2moW15nQgrq3GxWk,9860
111
- sglang/srt/models/qwen2.py,sha256=I2ZzH9pVTZdjP1fHlq1qdG4JiWHt1CC6t1EK2gN5Ppc,12337
112
- sglang/srt/models/qwen2_moe.py,sha256=BaNq8xgZKqjr_fcEBtH4yjBSc3-p4VztPiknVwllcQk,16984
113
- sglang/srt/models/qwen2_vl.py,sha256=9LLAaPFCzYVRfE2Nz3h3J-vvzU-do5jlyn0PUAXKupo,26304
114
- sglang/srt/models/stablelm.py,sha256=0NWUVsYGhbc_X2eT9x38MaaUhZGmFtMgw_2PBv25Yxw,11265
115
- sglang/srt/models/torch_native_llama.py,sha256=dtasdhwfRPE1eOcAIFUBsHrDnkjegXvo8WhGlqvXGKk,19154
116
- sglang/srt/models/xverse.py,sha256=v4OaFdss9oD5YNzXsnjoXE9ffCkXL9U5o0OWLm1vHQQ,13573
117
- sglang/srt/models/xverse_moe.py,sha256=A8EB82NpozoBplp7Qd8B_kY_3cL-UMydAxYIrhACVPE,15682
109
+ sglang/srt/models/mixtral.py,sha256=b6AqEgL4y9wQpKKAGnhgzBtUypYo3dca5UOpGNLEt_A,13949
110
+ sglang/srt/models/mixtral_quant.py,sha256=2ND-aOHjcyeQMUvqLLqhXwOdlR_bEftMFk3hc3lnpvc,13969
111
+ sglang/srt/models/mllama.py,sha256=pET1x8wY04yoS8HMCncKx0tFPqGp78K8rlA7Eq7XioE,37889
112
+ sglang/srt/models/olmo.py,sha256=eWPmo5AAnBhNGdMwklh1of3JnRzAszgQp4opeiiYidI,11887
113
+ sglang/srt/models/olmoe.py,sha256=fEWr-RmW6l6fVA8jM9KX8bumUWLNQQG8VxGpajlkhUs,15242
114
+ sglang/srt/models/qwen.py,sha256=vQoq8Bv8A2zc-LE1i-E97A8i4ydtfxb2yt2JG6Tp9PQ,9851
115
+ sglang/srt/models/qwen2.py,sha256=Y1f_PxZMTkSLgENbKl96VfNGBfvcU4cljpVe1a3vzVg,12328
116
+ sglang/srt/models/qwen2_moe.py,sha256=RRuHLN1fIYFS4du4pUPNzGL-Rt2wLrjlgDfXiczZQ5c,16975
117
+ sglang/srt/models/qwen2_vl.py,sha256=scKzs-KTI64CRRcBNWQniXURLO3WiJEzx-MsisH1Als,26093
118
+ sglang/srt/models/stablelm.py,sha256=rIQOv9OS_Vb2nOT_AMx0yGG2onwmCbbxvXL_SPdZX7k,11256
119
+ sglang/srt/models/torch_native_llama.py,sha256=d8gVNurlVVZ-tD3Uc_aHyGCVUUp1gR8awOH4fLRZHDE,19145
120
+ sglang/srt/models/xverse.py,sha256=meyCCdrZRYNK70hnmydgwhHa1FTBhKekEdpG0_IGTWY,13564
121
+ sglang/srt/models/xverse_moe.py,sha256=xlrhJBAlRzxhp5o0WQU_2V5Uvf8I9fwZLOZBh95o3to,15673
118
122
  sglang/srt/models/yivl.py,sha256=xcWqkuZ29FmBBJY6aKetwItWIPl-kfXK-QmgdLONles,4765
119
- sglang/srt/openai_api/adapter.py,sha256=WkYCKVaYTkFdLrySBhlkDyHJVaaHMF7KrhNnmw3L3us,53534
123
+ sglang/srt/openai_api/adapter.py,sha256=nZOVjZ-q4eULl19oT97_u7z63SQiWW7IzbYzJeWE7os,53069
120
124
  sglang/srt/openai_api/protocol.py,sha256=EZ6G209rBEDP7cepO2kAYqE8wMe1ksYdN7to1iT97Lw,10248
121
- sglang/srt/sampling/sampling_batch_info.py,sha256=dT81cFCe8zKor7P3g4ITnELTySX2nKX_FoNIU8y78Zo,7911
125
+ sglang/srt/sampling/sampling_batch_info.py,sha256=qrijXoMhF-V_x3g6lumsfXgsGaPiKLIJ2pUz6ii-O2s,7735
122
126
  sglang/srt/sampling/sampling_params.py,sha256=u1UWt9biIFXKymAg56RbkMa8oe5jxsoMvsv3cH7_kZ0,5692
123
127
  sglang/srt/sampling/penaltylib/__init__.py,sha256=5vQw0Y5DSzmsoFg1IdMIKLwFVhYZ5ArADHVBYbSmOec,513
124
128
  sglang/srt/sampling/penaltylib/orchestrator.py,sha256=kizcPnxtRawmDt6utRuhbk4yfNs5H5mx1DAlDVEZRv8,11328
@@ -129,7 +133,7 @@ sglang/srt/sampling/penaltylib/penalizers/repetition_penalty.py,sha256=v9jOgA0-I
129
133
  sglang/test/few_shot_gsm8k.py,sha256=ll-gNbcv829IwSPXAZt4JIEIu8IR3APCLcX3BHOFVp8,3968
130
134
  sglang/test/few_shot_gsm8k_engine.py,sha256=QQbrwOX6-cJDD3RZC_e7zPnt6aSo8JdF8X_lRHSjdDM,3886
131
135
  sglang/test/run_eval.py,sha256=9yO0hXZOcn4abEOs96T-XPguDEklK16Ltco0pGF3zCg,4020
132
- sglang/test/runners.py,sha256=zhv2CGKnJjhXNK6Y0U8uV_4g30QGtmO2KKxUohVgibw,14246
136
+ sglang/test/runners.py,sha256=JxfsGEW9L3cz87fHYmWqb3Vnbk6K1csLLLftR3LogxU,14297
133
137
  sglang/test/simple_eval_common.py,sha256=r0G-9QLycs2ax3RMc44T_61fzMxlpTzv6pececC7lyY,12379
134
138
  sglang/test/simple_eval_gpqa.py,sha256=8Xt9Bw05c7SZTYrCZgB68OZUqUbLo69ywiyx0bTvSUk,3220
135
139
  sglang/test/simple_eval_humaneval.py,sha256=7lTi841NT58smNOtRwCedrdX9IWWypdLkOtaQOBy-GI,5687
@@ -139,10 +143,10 @@ sglang/test/simple_eval_mmlu.py,sha256=FkwamjGMjueTixymkedF-YiPloSLiy4ftILFUrKZ9
139
143
  sglang/test/test_activation.py,sha256=jkdNRzJnbd5OgZliQaIXpxovlcky17UrweomcOcMxoE,1442
140
144
  sglang/test/test_layernorm.py,sha256=IacByD5d-stXjzBz8Ypamc7povlcedpKPbb_4JLgo3c,3720
141
145
  sglang/test/test_programs.py,sha256=1Z0umrsUu9pagzyGH5SrXl_qhKSyTfUv_kWC2mcn0qo,18208
142
- sglang/test/test_utils.py,sha256=NXbw0qyv0OpWMHAT0g_RDYiEYcjhTf1Ud5m-gRrX4kY,21234
146
+ sglang/test/test_utils.py,sha256=zspkM9VSm6QXI4wVG-75r8ttGgylnPOEH7nuYjp5plU,22799
143
147
  sglang/test/srt/sampling/penaltylib/utils.py,sha256=q98pQDikkmvvvvAG-AXMYaYte1iHHW2TFhKGtAeGvdE,12802
144
- sglang-0.3.4.post2.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
145
- sglang-0.3.4.post2.dist-info/METADATA,sha256=wC1SzkNFr9fGqhzFDqn6yRrjhIBgXtxsaEDYgekz0RQ,41431
146
- sglang-0.3.4.post2.dist-info/WHEEL,sha256=OVMc5UfuAQiSplgO0_WdW7vXVGAt9Hdd6qtN4HotdyA,91
147
- sglang-0.3.4.post2.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
148
- sglang-0.3.4.post2.dist-info/RECORD,,
148
+ sglang-0.3.5.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
149
+ sglang-0.3.5.dist-info/METADATA,sha256=FQ8MBpLt6W0-43VhtuwEWgqomXaFwUumiBd6T8xPWG0,21099
150
+ sglang-0.3.5.dist-info/WHEEL,sha256=P9jw-gEje8ByB7_hXoICnHtVCrEwMQh-630tKvQWehc,91
151
+ sglang-0.3.5.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
152
+ sglang-0.3.5.dist-info/RECORD,,
@@ -1,5 +1,5 @@
1
1
  Wheel-Version: 1.0
2
- Generator: setuptools (75.2.0)
2
+ Generator: setuptools (75.3.0)
3
3
  Root-Is-Purelib: true
4
4
  Tag: py3-none-any
5
5