sglang 0.4.10__py3-none-any.whl → 0.4.10.post2__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_offline_throughput.py +20 -0
- sglang/compile_deep_gemm.py +8 -1
- sglang/global_config.py +5 -1
- sglang/srt/configs/model_config.py +1 -0
- sglang/srt/conversation.py +0 -112
- sglang/srt/disaggregation/decode_schedule_batch_mixin.py +1 -0
- sglang/srt/disaggregation/launch_lb.py +5 -20
- sglang/srt/disaggregation/mooncake/conn.py +33 -15
- sglang/srt/disaggregation/prefill.py +1 -0
- sglang/srt/distributed/device_communicators/pynccl.py +7 -0
- sglang/srt/distributed/device_communicators/pynccl_allocator.py +133 -0
- sglang/srt/distributed/device_communicators/pynccl_wrapper.py +42 -3
- sglang/srt/distributed/parallel_state.py +11 -0
- sglang/srt/entrypoints/engine.py +4 -2
- sglang/srt/entrypoints/http_server.py +35 -15
- sglang/srt/eplb/expert_distribution.py +4 -2
- sglang/srt/hf_transformers_utils.py +25 -10
- sglang/srt/layers/attention/cutlass_mla_backend.py +3 -3
- sglang/srt/layers/attention/flashattention_backend.py +7 -11
- sglang/srt/layers/attention/trtllm_mla_backend.py +372 -0
- sglang/srt/layers/attention/utils.py +6 -1
- sglang/srt/layers/attention/vision.py +27 -10
- sglang/srt/layers/communicator.py +14 -4
- sglang/srt/layers/linear.py +7 -1
- sglang/srt/layers/logits_processor.py +9 -1
- sglang/srt/layers/moe/ep_moe/layer.py +29 -68
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=352,device_name=NVIDIA_RTX_6000_Ada_Generation,dtype=fp8_w8a8.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/layer.py +82 -25
- sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py +0 -31
- sglang/srt/layers/moe/token_dispatcher/__init__.py +23 -0
- sglang/srt/layers/moe/token_dispatcher/base_dispatcher.py +12 -1
- sglang/srt/layers/moe/{ep_moe/token_dispatcher.py → token_dispatcher/deepep.py} +8 -15
- sglang/srt/layers/moe/utils.py +43 -0
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py +3 -2
- sglang/srt/layers/quantization/deep_gemm_wrapper/compile_utils.py +1 -1
- sglang/srt/layers/quantization/fp8.py +57 -1
- sglang/srt/layers/quantization/fp8_kernel.py +0 -4
- sglang/srt/layers/quantization/w8a8_int8.py +4 -1
- sglang/srt/layers/vocab_parallel_embedding.py +7 -1
- sglang/srt/lora/lora_registry.py +7 -0
- sglang/srt/managers/cache_controller.py +43 -39
- sglang/srt/managers/data_parallel_controller.py +52 -2
- sglang/srt/managers/io_struct.py +6 -1
- sglang/srt/managers/schedule_batch.py +3 -2
- sglang/srt/managers/schedule_policy.py +3 -1
- sglang/srt/managers/scheduler.py +145 -6
- sglang/srt/managers/template_manager.py +25 -22
- sglang/srt/managers/tokenizer_manager.py +114 -62
- sglang/srt/managers/utils.py +45 -1
- sglang/srt/mem_cache/cpp_radix_tree/radix_tree.py +182 -0
- sglang/srt/mem_cache/hicache_storage.py +13 -12
- sglang/srt/mem_cache/hiradix_cache.py +21 -4
- sglang/srt/mem_cache/memory_pool.py +15 -118
- sglang/srt/mem_cache/memory_pool_host.py +350 -33
- sglang/srt/mem_cache/radix_cache_cpp.py +229 -0
- sglang/srt/mem_cache/storage/hf3fs/client_hf3fs.py +8 -2
- sglang/srt/mem_cache/storage/hf3fs/hf3fs_utils.cpp +35 -0
- sglang/srt/mem_cache/storage/nixl/hicache_nixl.py +163 -0
- sglang/srt/mem_cache/storage/nixl/nixl_utils.py +238 -0
- sglang/srt/mem_cache/storage/nixl/test_hicache_nixl_storage.py +216 -0
- sglang/srt/model_executor/cuda_graph_runner.py +42 -4
- sglang/srt/model_executor/forward_batch_info.py +13 -3
- sglang/srt/model_executor/model_runner.py +13 -1
- sglang/srt/model_loader/weight_utils.py +2 -0
- sglang/srt/models/deepseek_v2.py +28 -23
- sglang/srt/models/glm4_moe.py +85 -22
- sglang/srt/models/grok.py +3 -3
- sglang/srt/models/llama4.py +13 -2
- sglang/srt/models/mixtral.py +3 -3
- sglang/srt/models/mllama4.py +428 -19
- sglang/srt/models/qwen2_moe.py +1 -4
- sglang/srt/models/qwen3_moe.py +7 -8
- sglang/srt/models/step3_vl.py +1 -4
- sglang/srt/multimodal/processors/base_processor.py +4 -3
- sglang/srt/multimodal/processors/gemma3n.py +0 -7
- sglang/srt/operations_strategy.py +1 -1
- sglang/srt/server_args.py +115 -21
- sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py +18 -0
- sglang/srt/two_batch_overlap.py +6 -4
- sglang/srt/utils.py +4 -24
- sglang/srt/weight_sync/utils.py +1 -1
- sglang/test/attention/test_trtllm_mla_backend.py +945 -0
- sglang/test/runners.py +2 -2
- sglang/test/test_utils.py +3 -3
- sglang/version.py +1 -1
- {sglang-0.4.10.dist-info → sglang-0.4.10.post2.dist-info}/METADATA +3 -2
- {sglang-0.4.10.dist-info → sglang-0.4.10.post2.dist-info}/RECORD +92 -81
- /sglang/srt/mem_cache/{mooncake_store → storage/mooncake_store}/mooncake_store.py +0 -0
- /sglang/srt/mem_cache/{mooncake_store → storage/mooncake_store}/unit_test.py +0 -0
- {sglang-0.4.10.dist-info → sglang-0.4.10.post2.dist-info}/WHEEL +0 -0
- {sglang-0.4.10.dist-info → sglang-0.4.10.post2.dist-info}/licenses/LICENSE +0 -0
- {sglang-0.4.10.dist-info → sglang-0.4.10.post2.dist-info}/top_level.txt +0 -0
sglang/test/runners.py
CHANGED
@@ -499,7 +499,6 @@ class SRTRunner:
|
|
499
499
|
chunked_prefill_size: Optional[int] = None,
|
500
500
|
dp_size: int = 1,
|
501
501
|
tokenizer_path: Optional[str] = None,
|
502
|
-
enable_ep_moe: bool = False,
|
503
502
|
mem_fraction_static: float = 0.65,
|
504
503
|
trust_remote_code: bool = False,
|
505
504
|
speculative_draft_model_path: Optional[str] = None,
|
@@ -515,6 +514,7 @@ class SRTRunner:
|
|
515
514
|
max_lora_rank: Optional[int] = None,
|
516
515
|
lora_target_modules: Optional[List[str]] = None,
|
517
516
|
enable_lora: Optional[bool] = None,
|
517
|
+
max_loaded_loras: Optional[int] = None,
|
518
518
|
):
|
519
519
|
self.model_type = model_type
|
520
520
|
self.is_generation = model_type == "generation"
|
@@ -550,7 +550,6 @@ class SRTRunner:
|
|
550
550
|
enable_dp_attention=enable_dp_attention,
|
551
551
|
dp_size=dp_size,
|
552
552
|
tokenizer_path=tokenizer_path,
|
553
|
-
enable_ep_moe=enable_ep_moe,
|
554
553
|
disable_overlap_schedule=disable_overlap_schedule,
|
555
554
|
cuda_graph_max_bs=cuda_graph_max_bs,
|
556
555
|
disable_custom_all_reduce=disable_custom_all_reduce,
|
@@ -558,6 +557,7 @@ class SRTRunner:
|
|
558
557
|
max_lora_rank=max_lora_rank,
|
559
558
|
lora_target_modules=lora_target_modules,
|
560
559
|
enable_lora=enable_lora,
|
560
|
+
max_loaded_loras=max_loaded_loras,
|
561
561
|
**spec_kwargs,
|
562
562
|
)
|
563
563
|
|
sglang/test/test_utils.py
CHANGED
@@ -27,9 +27,6 @@ import torch.nn.functional as F
|
|
27
27
|
|
28
28
|
from sglang.bench_serving import run_benchmark
|
29
29
|
from sglang.global_config import global_config
|
30
|
-
from sglang.lang.backend.openai import OpenAI
|
31
|
-
from sglang.lang.backend.runtime_endpoint import RuntimeEndpoint
|
32
|
-
from sglang.lang.interpreter import ProgramState
|
33
30
|
from sglang.srt.utils import (
|
34
31
|
get_bool_env_var,
|
35
32
|
get_device,
|
@@ -358,6 +355,9 @@ def add_common_sglang_args_and_parse(parser: argparse.ArgumentParser):
|
|
358
355
|
|
359
356
|
|
360
357
|
def select_sglang_backend(args: argparse.Namespace):
|
358
|
+
from sglang.lang.backend.openai import OpenAI
|
359
|
+
from sglang.lang.backend.runtime_endpoint import RuntimeEndpoint
|
360
|
+
|
361
361
|
if args.backend.startswith("srt"):
|
362
362
|
if args.backend == "srt-no-parallel":
|
363
363
|
global_config.enable_parallel_encoding = False
|
sglang/version.py
CHANGED
@@ -1 +1 @@
|
|
1
|
-
__version__ = "0.4.10"
|
1
|
+
__version__ = "0.4.10.post2"
|
@@ -1,6 +1,6 @@
|
|
1
1
|
Metadata-Version: 2.4
|
2
2
|
Name: sglang
|
3
|
-
Version: 0.4.10
|
3
|
+
Version: 0.4.10.post2
|
4
4
|
Summary: SGLang is yet another fast serving framework for large language models and vision language models.
|
5
5
|
License: Apache License
|
6
6
|
Version 2.0, January 2004
|
@@ -250,7 +250,7 @@ Requires-Dist: transformers==4.54.1; extra == "runtime-common"
|
|
250
250
|
Requires-Dist: timm==1.0.16; extra == "runtime-common"
|
251
251
|
Requires-Dist: uvicorn; extra == "runtime-common"
|
252
252
|
Requires-Dist: uvloop; extra == "runtime-common"
|
253
|
-
Requires-Dist: xgrammar==0.1.
|
253
|
+
Requires-Dist: xgrammar==0.1.22; extra == "runtime-common"
|
254
254
|
Provides-Extra: srt
|
255
255
|
Requires-Dist: sglang[runtime_common]; extra == "srt"
|
256
256
|
Requires-Dist: sgl-kernel==0.2.8; extra == "srt"
|
@@ -301,6 +301,7 @@ Requires-Dist: matplotlib; extra == "test"
|
|
301
301
|
Requires-Dist: pandas; extra == "test"
|
302
302
|
Requires-Dist: peft; extra == "test"
|
303
303
|
Requires-Dist: sentence_transformers; extra == "test"
|
304
|
+
Requires-Dist: pytest; extra == "test"
|
304
305
|
Provides-Extra: all
|
305
306
|
Requires-Dist: sglang[srt]; extra == "all"
|
306
307
|
Requires-Dist: sglang[openai]; extra == "all"
|
@@ -1,16 +1,16 @@
|
|
1
1
|
sglang/__init__.py,sha256=x1v8748WXS2BAptWAd5fR3YBEkngC7e_P_rkCNuNCGg,1702
|
2
2
|
sglang/api.py,sha256=rcp3GeoyZhmJ0GDLPRkuZNcxd0TBJy_wfUDpcmQoqW8,7210
|
3
|
-
sglang/bench_offline_throughput.py,sha256=
|
3
|
+
sglang/bench_offline_throughput.py,sha256=Bge0KG4Uiv9CHsxDnqKdXJmG_TcxM4blz3CmvLmDRvM,14815
|
4
4
|
sglang/bench_one_batch.py,sha256=kSqUxn8kEib6GkEdSDZgAaY6P1JuFL2BadbfVuXJjQs,19610
|
5
5
|
sglang/bench_one_batch_server.py,sha256=LS1BRmFwP67IpYlU5wmkWjqquiA5drvtWe-fwHnFSKI,14170
|
6
6
|
sglang/bench_serving.py,sha256=nOjDnqOKDezDvtCvEhC1_FAoWVkir38bSaEIcSGGVZg,71748
|
7
7
|
sglang/check_env.py,sha256=qDMIG2rCNBH1yKnxQmF-Bp10oiFMUKMgfZLHZYOmdSY,8412
|
8
|
-
sglang/compile_deep_gemm.py,sha256=
|
9
|
-
sglang/global_config.py,sha256=
|
8
|
+
sglang/compile_deep_gemm.py,sha256=EsgboGA-MK3Rvx9TlUJPLxJ5LBira4bTcHJLKEp1H6k,6488
|
9
|
+
sglang/global_config.py,sha256=ZMTux_PsGnvkyJ0kTFwhTdbnFwIjjpGDogut_9Lu4Vo,1732
|
10
10
|
sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
|
11
11
|
sglang/profiler.py,sha256=tEHzHerXC-ymk4OrkoUcMbgcGHmb8VESthsNSP2Yx9w,4417
|
12
12
|
sglang/utils.py,sha256=C_r3OF-OA6DwUz6KYQF6xWUBApjHaqSPsV9JccafrWI,16403
|
13
|
-
sglang/version.py,sha256=
|
13
|
+
sglang/version.py,sha256=Ypmiam91zyRF3ijHnORC1GWsK_sC2iafbARSRi4vW8Y,29
|
14
14
|
sglang/eval/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
|
15
15
|
sglang/eval/loogle_eval.py,sha256=-CC2s2kh5qUoDrHRkQVkC_jNvBgNojXbf456ny5s78s,4557
|
16
16
|
sglang/lang/chat_template.py,sha256=ZNcuDvDZE9781tQqfKpptck_zJVK0aggq7IxrNHltao,21934
|
@@ -30,20 +30,20 @@ sglang/srt/_custom_ops.py,sha256=9GPNUyqnybgvUnzdQ8n5_NgluHmQNw0Gptos0iLPfrY,534
|
|
30
30
|
sglang/srt/aio_rwlock.py,sha256=6LYtOdeTUY3hkfa1dmYkgsaF2ttrwIF3hUWz2AZ2fqw,2970
|
31
31
|
sglang/srt/code_completion_parser.py,sha256=KFa95OU0TeVjJkOqIgS4xV3kaJ5dFWOmAAgISyc1oEc,3803
|
32
32
|
sglang/srt/constants.py,sha256=0i-tEwG2BSYNDy96MxnGHV5HnBELkYcnsVGsE-R18o0,93
|
33
|
-
sglang/srt/conversation.py,sha256=
|
33
|
+
sglang/srt/conversation.py,sha256=0lMeOqBiVhmIwNjt9TxwyuraNk6XGpYKCpq4CLB5ai4,38730
|
34
34
|
sglang/srt/custom_op.py,sha256=7OPQAZe3PqDXk-xM0Wxl7pm9cazilNJfkNiYIWztWgI,3198
|
35
|
-
sglang/srt/hf_transformers_utils.py,sha256
|
35
|
+
sglang/srt/hf_transformers_utils.py,sha256=dSMTGS1Cac25-5Ygj2rUZba78VFnXM0acyMAW5aU8NY,13059
|
36
36
|
sglang/srt/jinja_template_utils.py,sha256=LHqdxzuYfUbRY-QV8tYYPejulf1RrxNy--M3JFLKEZU,7255
|
37
37
|
sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
|
38
38
|
sglang/srt/operations.py,sha256=ddQ8KO63L73OciaR8MZ9h2h83gKVY4-WuWgeEGowPJA,5346
|
39
|
-
sglang/srt/operations_strategy.py,sha256
|
39
|
+
sglang/srt/operations_strategy.py,sha256=-w6GI2VgSGvORy3pUWRAJo6244ykSW1ZBV1Dx0fZlHA,7138
|
40
40
|
sglang/srt/patch_torch.py,sha256=OUPCGQSQz3MVZB1zZ_Eq8lXiw0uIKJ_HWjqQolI8FsM,3088
|
41
41
|
sglang/srt/poll_based_barrier.py,sha256=9vjn32ROlSueEk7EuZjWKUrrL7XVR9KJsSVliGTYB_4,937
|
42
42
|
sglang/srt/reasoning_parser.py,sha256=pN7SbLjaiyT5tdrAZgOFWy3QrDKnZEHkBFeIKFXknZM,9288
|
43
|
-
sglang/srt/server_args.py,sha256=
|
43
|
+
sglang/srt/server_args.py,sha256=3DwQscmXEUNALT294DdVuMhXqjfA6CXKUqOEq-vXGxY,90227
|
44
44
|
sglang/srt/torch_memory_saver_adapter.py,sha256=K_eTx0UU84MHSTXI3iqYLdHV4IWtJMJ2FKdGFJR8v1E,2417
|
45
|
-
sglang/srt/two_batch_overlap.py,sha256=
|
46
|
-
sglang/srt/utils.py,sha256=
|
45
|
+
sglang/srt/two_batch_overlap.py,sha256=X5DE7EcOdr1XQQLzrZwzrfahqYzdBSvBjhnkfIClpEk,28964
|
46
|
+
sglang/srt/utils.py,sha256=ILvrvwyiV6iCOVEJ7W331q4jTGzysD8hMl8ao50EVFY,92832
|
47
47
|
sglang/srt/warmup.py,sha256=zldxhMlXpclRAJXmfBjJNUJd1eDizVdysibBvQyTVuA,1782
|
48
48
|
sglang/srt/configs/__init__.py,sha256=gas3cKYVgZp7qF4m8FejX4EUUmdQd_w5WDTm7fo7yM0,743
|
49
49
|
sglang/srt/configs/chatglm.py,sha256=j-b0YkdYUmQm2y1kNmMJtKeACxWKmBbvNNkDWbs6kbI,2907
|
@@ -56,7 +56,7 @@ sglang/srt/configs/janus_pro.py,sha256=yI2c_aWBIKiTcK6RZxK6hq8CIxyYRH_NN8uaSNKsr
|
|
56
56
|
sglang/srt/configs/kimi_vl.py,sha256=4W7VQI3pr888ZsFA2SqCQo4mI0seXTOrGQ-x3oTvWew,1358
|
57
57
|
sglang/srt/configs/kimi_vl_moonvit.py,sha256=hx2Rt4JSFbvy2HUTeLjBpge87m8M6ITAhqsgdNf_Jd4,1163
|
58
58
|
sglang/srt/configs/load_config.py,sha256=qs-AxuplouBx2tsv9KGBOLZPbwzuVA4vbktbGP_cRp8,3309
|
59
|
-
sglang/srt/configs/model_config.py,sha256=
|
59
|
+
sglang/srt/configs/model_config.py,sha256=opeGqkpJcCjPLNLSa_Wf2PfQPB17leJKYO_VgYi17q8,28888
|
60
60
|
sglang/srt/configs/step3_vl.py,sha256=_Otgnym57DVgB_kZ__8c1_Ys5gSalA_K0ZuVjcG51T0,4845
|
61
61
|
sglang/srt/configs/update_config.py,sha256=D-myV9Ombwp6wTF0RX9qUO6m5rxFiUcyse32FQADG28,4787
|
62
62
|
sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
|
@@ -80,11 +80,11 @@ sglang/srt/debug_utils/dump_comparator.py,sha256=8ZPwP0zojfRHvPiSKgpESYFcC3tR-sB
|
|
80
80
|
sglang/srt/debug_utils/dumper.py,sha256=eyrg5lmwyYF88kn8b06kpu1__aGnWb6CLM8fxywtooI,3158
|
81
81
|
sglang/srt/debug_utils/text_comparator.py,sha256=pg58q3fzsnd3rJKRgZTdcOviOtAv8pSYZdwf-TLffMM,5522
|
82
82
|
sglang/srt/disaggregation/decode.py,sha256=xPXKHz0e8O9GZetHiGVQoN8iIwEA1GwO8jj0NkCzgp4,33884
|
83
|
-
sglang/srt/disaggregation/decode_schedule_batch_mixin.py,sha256=
|
83
|
+
sglang/srt/disaggregation/decode_schedule_batch_mixin.py,sha256=R3JmjHHU_j6aOtUkilHsL5hzNn5oJpZUqmX_mEsT0Ek,6485
|
84
84
|
sglang/srt/disaggregation/kv_events.py,sha256=yFbtwOeblcCmOXTjg00TopxnyjkmCBQIVz46KB8jetY,13555
|
85
|
-
sglang/srt/disaggregation/launch_lb.py,sha256=
|
85
|
+
sglang/srt/disaggregation/launch_lb.py,sha256=Elw-J-97jTA70Ml1EMp_J-k9X9H5WHTWOC2XEvmyJRY,3827
|
86
86
|
sglang/srt/disaggregation/mini_lb.py,sha256=bpnz6d90nv7vAoiPxmxFKlM1Gjct0R4IPkF9SwwBk9I,14504
|
87
|
-
sglang/srt/disaggregation/prefill.py,sha256=
|
87
|
+
sglang/srt/disaggregation/prefill.py,sha256=03li8dKrOFfhUsvaSuGKQWgH9eAyWaYv59cBiWpzo-s,24136
|
88
88
|
sglang/srt/disaggregation/utils.py,sha256=ql5p9yjBCRcR0YMgcgRK6tH0kPGvLNTUgeXBr_BUcoI,12198
|
89
89
|
sglang/srt/disaggregation/ascend/__init__.py,sha256=-lxnica6fZYmYNpKwmE8yESUpQ5Sxf2DiZoBHzboKc8,146
|
90
90
|
sglang/srt/disaggregation/ascend/conn.py,sha256=FOYrGtEhj58iTjhYT1eP96DpyxBLFFR87icqYCzWmCY,1215
|
@@ -97,13 +97,13 @@ sglang/srt/disaggregation/common/utils.py,sha256=SxRhAWisNK8seGhb5BXBJ5u53DF7yeK
|
|
97
97
|
sglang/srt/disaggregation/fake/__init__.py,sha256=jJGWdXwaQiGIoR6atKqkQfkJmVyQ09l55VUN2WjwaeY,77
|
98
98
|
sglang/srt/disaggregation/fake/conn.py,sha256=oD1DArn1yDFZCu-X6p93uSLlAXEkt9lYxERICMznxGw,2286
|
99
99
|
sglang/srt/disaggregation/mooncake/__init__.py,sha256=0TgqkAdQI1YynbHY6c0QISvVoOSk-0SwCIq5rjPSmgE,156
|
100
|
-
sglang/srt/disaggregation/mooncake/conn.py,sha256=
|
100
|
+
sglang/srt/disaggregation/mooncake/conn.py,sha256=BrOcdnqk7Ut90qDqajNzQx3qTBlnsVhLmMgUgXVGCdg,62677
|
101
101
|
sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=-ulLhz6DBDsmEjwNjjH5Pj8ngZKISgwBc_zL3ViGDN0,4761
|
102
102
|
sglang/srt/disaggregation/nixl/__init__.py,sha256=qODVPIGWUXKXq4zsRIcMYoAoAeg6nBIN9vdQOlVMANE,136
|
103
103
|
sglang/srt/disaggregation/nixl/conn.py,sha256=eSof87fG21Dd4COszfnbeXIxne3TWvw0mSvCOkjsBZc,20323
|
104
104
|
sglang/srt/distributed/__init__.py,sha256=jFOcyt-wFAPMBUAf9zkZalNQlt-4rqmT6pCKBz1E4qo,149
|
105
105
|
sglang/srt/distributed/communication_op.py,sha256=IBnFUdMftK_VSTMMMitGveonorFUUVNL4guqO31cMSc,1130
|
106
|
-
sglang/srt/distributed/parallel_state.py,sha256=
|
106
|
+
sglang/srt/distributed/parallel_state.py,sha256=NAIEJWT5IgMqz-Kpb0SKgok2HElfXWWTOABUqx-t53Q,61675
|
107
107
|
sglang/srt/distributed/utils.py,sha256=aaCxATncLGnVgB0WlGpBdee0behKW8Dy_dakqcuKSaQ,8497
|
108
108
|
sglang/srt/distributed/device_communicators/cuda_wrapper.py,sha256=3jvPG-Ow5UBLiXhfx8T8snR7crSZbPpARAggsDPWq7k,7038
|
109
109
|
sglang/srt/distributed/device_communicators/custom_all_reduce.py,sha256=p-GLYYnMSjcUvK30qcwYhXEJwGrOz6rR99p_23SWFj8,16570
|
@@ -111,14 +111,15 @@ sglang/srt/distributed/device_communicators/custom_all_reduce_utils.py,sha256=Fx
|
|
111
111
|
sglang/srt/distributed/device_communicators/hpu_communicator.py,sha256=gPjEH1-izoby5uDrfUlzNf21luPT0Ow7pJjhCRKnHy8,1728
|
112
112
|
sglang/srt/distributed/device_communicators/npu_communicator.py,sha256=bRXN1Md_4SHQGzQYZa2GrHv2zbIU5vSpkueHiAZL1xQ,1345
|
113
113
|
sglang/srt/distributed/device_communicators/pymscclpp.py,sha256=8Pgehd02v-BpHixTTB4OB9ZlxA7fyXiPF4Xp9F_heyU,10890
|
114
|
-
sglang/srt/distributed/device_communicators/pynccl.py,sha256=
|
115
|
-
sglang/srt/distributed/device_communicators/
|
114
|
+
sglang/srt/distributed/device_communicators/pynccl.py,sha256=_p60Ow-WNBBsjxUCVaiFd7TBLuFpgZRW0PnI_ImrsX0,10417
|
115
|
+
sglang/srt/distributed/device_communicators/pynccl_allocator.py,sha256=zAxWwI73ZNsF_c4PZUnRMzwzSD88v26w8komzZ4Uatc,4771
|
116
|
+
sglang/srt/distributed/device_communicators/pynccl_wrapper.py,sha256=C1cwXmjgnWH6Eu1zaq4zWqsiYGu0H20sUlwSF4n7KhI,16833
|
116
117
|
sglang/srt/distributed/device_communicators/quick_all_reduce.py,sha256=4j1_E4azoxfd8wxtfFmt9rvbQncl8ny6wmTMl6gAkp0,9932
|
117
118
|
sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=IrSrnpZnii0EJTS1CYRwEwE7gyHxJBVgI2QuJS3AKW0,20906
|
118
119
|
sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
|
119
120
|
sglang/srt/entrypoints/EngineBase.py,sha256=yKN76witT2jz1zhmLHmPNLGMpK2UiOTaKQ2KPD8l99U,2594
|
120
|
-
sglang/srt/entrypoints/engine.py,sha256=
|
121
|
-
sglang/srt/entrypoints/http_server.py,sha256=
|
121
|
+
sglang/srt/entrypoints/engine.py,sha256=P0K6pCyznxDheUVmzLhmQPTX4arVPQY0PAGQKQhwZNI,31856
|
122
|
+
sglang/srt/entrypoints/http_server.py,sha256=qdi-Qf3UQnufdRMx4m-MaBsN1maJ3oN3VQldiNFyOec,39069
|
122
123
|
sglang/srt/entrypoints/http_server_engine.py,sha256=_--j4U04OeJLlnnv1f0XmCd_Ry0z1FlhkrbePX8rYV0,4938
|
123
124
|
sglang/srt/entrypoints/openai/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
124
125
|
sglang/srt/entrypoints/openai/protocol.py,sha256=2XfURpoICsk4UZs4W9lm2A1yXNC-iwqSdZOuof0FXjo,19183
|
@@ -132,7 +133,7 @@ sglang/srt/entrypoints/openai/usage_processor.py,sha256=9LTB5rqdRuMKyZrIXiUBuF_W
|
|
132
133
|
sglang/srt/entrypoints/openai/utils.py,sha256=DT9quW0F1DX1P9NAuMxpERgXcNYKKtm7rd7J1m7EaDQ,2112
|
133
134
|
sglang/srt/eplb/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
134
135
|
sglang/srt/eplb/eplb_manager.py,sha256=Wt7w4L5g1BSsXx1mqoGo8WjkB55f4nGB4KqC3y_ijl8,3394
|
135
|
-
sglang/srt/eplb/expert_distribution.py,sha256=
|
136
|
+
sglang/srt/eplb/expert_distribution.py,sha256=PG40hrCP2-RJR6hmS80I0N5HUj2XZnxR0CBBFxq9xvg,31740
|
136
137
|
sglang/srt/eplb/expert_location.py,sha256=A9zkr4lCl8480WlPVIMosdjSrMI1y_ym_v1iRIjJP68,16753
|
137
138
|
sglang/srt/eplb/expert_location_dispatch.py,sha256=vA0o-7bXf847BQAVCqElWKEpofZLAIsir_d0pew2KNQ,4178
|
138
139
|
sglang/srt/eplb/expert_location_updater.py,sha256=eMtJVMCRklfvKPbFiDKy85f5J9s_krJbpRCSArYZjVs,21157
|
@@ -157,13 +158,13 @@ sglang/srt/function_call/step3_detector.py,sha256=SAibrispu0peDQzqOG2-V93KGHtV6U
|
|
157
158
|
sglang/srt/function_call/utils.py,sha256=__ImDF2kNyoLWsYO5RYoryvy1mmgEjnjXlCvLv-uLCM,1695
|
158
159
|
sglang/srt/layers/activation.py,sha256=uhfhh10N5iLoLEBCvFPRjCfDxu5jXe1ehHJb3ISIjig,7840
|
159
160
|
sglang/srt/layers/amx_utils.py,sha256=1mENgHK2B8mgaD1oMtgbZ15Jmy_Uu1QueBmo09Ff2iA,2865
|
160
|
-
sglang/srt/layers/communicator.py,sha256=
|
161
|
+
sglang/srt/layers/communicator.py,sha256=xzVeuW2ymLERcKgqVNLl8Hv7hBr0_scc-b4Vx1-wpOA,20453
|
161
162
|
sglang/srt/layers/dp_attention.py,sha256=tOre7il5Cppu930-tzM6d4AgjHuIJMCMlg9hJkuBcjA,11382
|
162
163
|
sglang/srt/layers/elementwise.py,sha256=MyQUflyKEfPZ-BggW1Kd4hB53RFD6FXGc2S5LXjx_do,16026
|
163
164
|
sglang/srt/layers/flashinfer_comm_fusion.py,sha256=fkTcAB7qYwSWi95qI3Rqq0JUyDpJdcYkd4TYkWO01HI,5891
|
164
165
|
sglang/srt/layers/layernorm.py,sha256=Qkx4Jzb3KWOt2okAVM6Is4FDClpUIJShcPGoElUSnEQ,9598
|
165
|
-
sglang/srt/layers/linear.py,sha256=
|
166
|
-
sglang/srt/layers/logits_processor.py,sha256=
|
166
|
+
sglang/srt/layers/linear.py,sha256=Fp5xLw1kCkj-zpRc9_YH4B7TNkYcG4E0WURkzc-Fbwo,52706
|
167
|
+
sglang/srt/layers/logits_processor.py,sha256=LToJR70CfI7_gBLVNLmn5kUTovj3BU_7UDc4hjlac08,27077
|
167
168
|
sglang/srt/layers/multimodal.py,sha256=wVo5-OLz-J4k8fM0ld1ZKMtzOsSue4aa-uILak-I8Bk,2398
|
168
169
|
sglang/srt/layers/parameter.py,sha256=jCg0G-12GZqTa9gGOqEtVCsnnuBGnYg0VmF6pc6oCFg,17455
|
169
170
|
sglang/srt/layers/pooler.py,sha256=uZ6WX1FLMEafZwusyZdm6KuVlIwSjbKrdwk2qzgqNGk,3812
|
@@ -172,13 +173,13 @@ sglang/srt/layers/rotary_embedding.py,sha256=EhxI0E8jcTWZ2COpnku7crbW8Hew5fe_ujM
|
|
172
173
|
sglang/srt/layers/sampler.py,sha256=xNds1migup2s6b9_pS6ljkJUkvNtv7nmTGeIdOzoQ6w,11182
|
173
174
|
sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSahyMfA,4063
|
174
175
|
sglang/srt/layers/utils.py,sha256=IWGg1Hb7c33Z3LHRPVJyUAzp3BnSid23ZWXAmJ_Jvp8,1204
|
175
|
-
sglang/srt/layers/vocab_parallel_embedding.py,sha256=
|
176
|
+
sglang/srt/layers/vocab_parallel_embedding.py,sha256=UT0GFg94Kwvs9RFRdKR9ISBsftcXycP7s_nZQnf_urs,22590
|
176
177
|
sglang/srt/layers/attention/aiter_backend.py,sha256=7sEUgViw-xl3yok91yyOD9gTi8lQmME0g0ZiKVTCcyI,32851
|
177
178
|
sglang/srt/layers/attention/ascend_backend.py,sha256=jPCsU9_gH1iZNoZHD9nCeDdVdXqBt31LI65N55BTJPg,8250
|
178
179
|
sglang/srt/layers/attention/base_attn_backend.py,sha256=_vM05ddWy6SaoqWYPR8w5lflSE4MhcVFR271Q9EKK8E,3600
|
179
|
-
sglang/srt/layers/attention/cutlass_mla_backend.py,sha256=
|
180
|
+
sglang/srt/layers/attention/cutlass_mla_backend.py,sha256=PB5BjbJHhmTg_t0kXfarTI6C3pPvKhBAqvCSGGajou0,9846
|
180
181
|
sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
|
181
|
-
sglang/srt/layers/attention/flashattention_backend.py,sha256=
|
182
|
+
sglang/srt/layers/attention/flashattention_backend.py,sha256=22vHnwDBDTtx0WoW76TsYUPewPu9OqCYQtJmLKqVJ2s,93917
|
182
183
|
sglang/srt/layers/attention/flashinfer_backend.py,sha256=HQp14R29hQVXYjATKPoGaHY9LJ9e1cw2k-RhBEmFxt8,51505
|
183
184
|
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=NirRlNFxD0S2EncvvmrerIxNfuTnYpvemZV3MR5_FnU,34040
|
184
185
|
sglang/srt/layers/attention/flashmla_backend.py,sha256=5iSic5ho-lkXNas9mR3uLbXbEl-do31gc8gjR7-a79k,20711
|
@@ -188,8 +189,9 @@ sglang/srt/layers/attention/merge_state.py,sha256=OnWmNV6LYTUj0y8SCy_IK7A9q6v9bZ
|
|
188
189
|
sglang/srt/layers/attention/tbo_backend.py,sha256=CBnSDUmDvT-Tbhi9NHNo1LLkgUAcIsWMJgKa-sAIKW8,9318
|
189
190
|
sglang/srt/layers/attention/torch_native_backend.py,sha256=RsI4lYiQJAj-Xy0vKkzbyKOWHAVPOD9b0n0dEehwlos,9496
|
190
191
|
sglang/srt/layers/attention/triton_backend.py,sha256=AryrPc4kzioxkiucPkApMA__0HoQTKOHNpwN0GEgm98,36923
|
191
|
-
sglang/srt/layers/attention/
|
192
|
-
sglang/srt/layers/attention/
|
192
|
+
sglang/srt/layers/attention/trtllm_mla_backend.py,sha256=KwKvpnL6CpO84oDRuLVnGfNtT3SQ1SRYCZRI0ElGabA,13446
|
193
|
+
sglang/srt/layers/attention/utils.py,sha256=lx_jNw2VwlBe2VnDb1eXRRdN9WArw9rdWHTHqndUEFk,3174
|
194
|
+
sglang/srt/layers/attention/vision.py,sha256=8eRhnwpUW646j03pAhyswjpmHSpDWryv7foIoDI0cNs,18828
|
193
195
|
sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=eN0Lqpu5h27lMPuRnisX4jrU0WR061pCFSrh3vLlSD0,18936
|
194
196
|
sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=FbgrPO5GvEGEvojV7Ss0jX7384e9vuzX2goDCQuP1Zk,31124
|
195
197
|
sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=ImpCuxszoeVdr6wzaLR3OGW3-50P9EAEvsRCdUVIrxE,14261
|
@@ -202,14 +204,14 @@ sglang/srt/layers/moe/cutlass_w4a8_moe.py,sha256=bc8s5Oc36pCF0VozxZuqFOOU2fov-4w
|
|
202
204
|
sglang/srt/layers/moe/fused_moe_native.py,sha256=SkOFgU23aZfHmLhnkVhdUWGx0-2wbo6uu9ohKBp2yOI,3486
|
203
205
|
sglang/srt/layers/moe/router.py,sha256=UrPieRvemN7Ew48gtG7DA2xhNDBRSnZxzugTEBI-0_E,12006
|
204
206
|
sglang/srt/layers/moe/topk.py,sha256=ttPFvXzTY0eyh6UH8fJZCjyDp-R9EssSP8m229MxQ_4,26925
|
207
|
+
sglang/srt/layers/moe/utils.py,sha256=UcHdqZi57kgDRjF6nWnw7lZKGj877IVB4VaSK8qNtC4,1051
|
205
208
|
sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
206
209
|
sglang/srt/layers/moe/ep_moe/kernels.py,sha256=UFH-XacP3aBjAcn6_SeC641jGzjSrY0fz50IB9URix4,42706
|
207
|
-
sglang/srt/layers/moe/ep_moe/layer.py,sha256=
|
208
|
-
sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=CJhFTtOydsLSXMYNfzAASPI5p_oP7c4krGwX4bBcT1s,22909
|
210
|
+
sglang/srt/layers/moe/ep_moe/layer.py,sha256=M4X5HOu5k0mumtzG2cif7OLuM08IKk8nyCDED08qKjA,26261
|
209
211
|
sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=4NXZHbCw-G-uSnNUj4up0yh3xBDPnT-x0pdoIr0lku8,831
|
210
212
|
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=Yb7LvdCQWonJC05_4UOMt3jQYWmRl-AkdY7H5EVwKhM,57575
|
211
|
-
sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=
|
212
|
-
sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=
|
213
|
+
sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=z7HD4oNe7fITnlmmFMS4L6gf4GcpS_9CLXJ-B5dAh7w,27874
|
214
|
+
sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=q8o4puG3ous6wTMbQFzydQ3P5emZbsNKepNaveLSpL8,4956
|
213
215
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=iNGsE2ZeVnQEnN4A8UJ9Jv0d3hbRF2MJ9oBgjup5Szk,2737
|
214
216
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=JJN0hryyLr5Zv3dSS7C8cPFhAwTT6XxUVnBGMZvV6JA,2752
|
215
217
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=ouRyZ5PEMPP2njPftCNhs-1g1y6wueWLmhI7G1SjV1k,4131
|
@@ -370,6 +372,7 @@ sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=HOEJJyehKTUI
|
|
370
372
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=272,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=Piw4LN6d8QYrUahWsw3XUOtTMD1o3vHPwA94sGI56Gk,3242
|
371
373
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=288,N=64,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=3T8_rF2PEojhgTMyQ8DscXgJCWWdWfDPj4M434zWcA4,3243
|
372
374
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=tme0ydWzIxdABZLk4tU8G_X2dJUYGGZNkQzNGcmcvUc,3261
|
375
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=352,device_name=NVIDIA_RTX_6000_Ada_Generation,dtype=fp8_w8a8.json",sha256=RMYwqtB7hYeaCyuZ1feVnRhBGlN4jMcmNQX2hqUqBlk,3254
|
373
376
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=384,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=LNx1W_BsWZcpHomiScCRap46dV0-F7S_w3Htskoqlm8,3263
|
374
377
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json",sha256=rqPWpRfa9s7USa9cUyHAVP50WYkCqXDJzfKZhqk6GHw,3267
|
375
378
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=320,device_name=NVIDIA_H20-3e.json",sha256=cWkEpNqnyn0QS8HcgWiwWI1xqh_U93_S1kyNeb69aOc,3238
|
@@ -381,16 +384,17 @@ sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=HOEJJyehKTUI
|
|
381
384
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=385,N=128,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=rgF-KSAUXmI7QGrhMUywPVfeTxz_j8_IP4N-6bkYdak,3260
|
382
385
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=385,N=128,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=7aaoi0MI2xC5ZJkflkTZAH4tawTPdmUpmcBOgOInKxs,3268
|
383
386
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=a9XJZ0XudUTebAdN5guVq-3XSvZetHIRjatO1-_0xCk,3257
|
384
|
-
sglang/srt/layers/moe/token_dispatcher/__init__.py,sha256=
|
385
|
-
sglang/srt/layers/moe/token_dispatcher/base_dispatcher.py,sha256=
|
387
|
+
sglang/srt/layers/moe/token_dispatcher/__init__.py,sha256=QUjIW1xImUsIf9M8Ze6gervzEdRvvMpgtt_xcYERALQ,516
|
388
|
+
sglang/srt/layers/moe/token_dispatcher/base_dispatcher.py,sha256=ibQnHlcnemt8mwOMj01qz360XaherXvTY0T6wGOeBQU,1273
|
389
|
+
sglang/srt/layers/moe/token_dispatcher/deepep.py,sha256=CbPI6Ul1GjrT81NMbhL2E_CSt6cwFFepMtmcEyTAv54,22830
|
386
390
|
sglang/srt/layers/moe/token_dispatcher/standard.py,sha256=jseBfZpst5iCDxXxQV_-53qb0VMqy8RHqXJMf86Bdgg,430
|
387
391
|
sglang/srt/layers/quantization/__init__.py,sha256=lMjOmhMntbY55RR5eKKLB338tdzlxGuBx-pT4f9jmIM,8130
|
388
392
|
sglang/srt/layers/quantization/awq.py,sha256=gLEOx6QoGQ0McBEURXPVidKGsdNUjqc9sJwGLQx86rE,26690
|
389
393
|
sglang/srt/layers/quantization/awq_triton.py,sha256=4XwJbRNlFmlvjbMGtWF1Wvk26Kf96162vNEmuI5Kars,11737
|
390
394
|
sglang/srt/layers/quantization/base_config.py,sha256=CiSxG5iuOa-12V9kIOh0Fu8Y7knZxfFbbNGTo6nEaRE,7173
|
391
395
|
sglang/srt/layers/quantization/blockwise_int8.py,sha256=3avetVUBjCiuYWBJQbElbVXWpx0SgsFsDc5F9F_zHgA,13879
|
392
|
-
sglang/srt/layers/quantization/fp8.py,sha256=
|
393
|
-
sglang/srt/layers/quantization/fp8_kernel.py,sha256=
|
396
|
+
sglang/srt/layers/quantization/fp8.py,sha256=zo7Uh5De-jk3Wowdha52oWl70uc9sfbJgdNSTRrwDP4,48815
|
397
|
+
sglang/srt/layers/quantization/fp8_kernel.py,sha256=0iC_rlmWVEfZOtr5osgMrd8-QsW2pCi-paaX1NqoVdU,40694
|
394
398
|
sglang/srt/layers/quantization/fp8_utils.py,sha256=6FVR6_Y8akP4ihCaGG_bLYurcC_PShPtiPgiEvGN5bE,25605
|
395
399
|
sglang/srt/layers/quantization/gptq.py,sha256=3q9xfq5sJJHcrVr9C8-l9JdcLFvwKAAfWJEfZW6CzLs,39145
|
396
400
|
sglang/srt/layers/quantization/int8_kernel.py,sha256=GfRC9FOn9exNvK4QHbUeBj3Hhv32VcyGphapFPt5b84,12625
|
@@ -407,10 +411,10 @@ sglang/srt/layers/quantization/unquant.py,sha256=vgI2G1iHpMr2ys1-HaQ_YIXJN0z1m9u
|
|
407
411
|
sglang/srt/layers/quantization/utils.py,sha256=9ZmfS0MMqueQ6zIxMIFoQy2OFBQ5vHjOHBgVvUn7ews,16213
|
408
412
|
sglang/srt/layers/quantization/w4afp8.py,sha256=wa5t0g2IwVFcwHxZ213f5Vsnt8P4fum7WVoxvf0llpw,11092
|
409
413
|
sglang/srt/layers/quantization/w8a8_fp8.py,sha256=I_W9A4Pnluaf-SaT8zq-tAJTt1acpxfE_kG5mqdc5cE,10250
|
410
|
-
sglang/srt/layers/quantization/w8a8_int8.py,sha256=
|
414
|
+
sglang/srt/layers/quantization/w8a8_int8.py,sha256=VXGAMKMlRU7k6c-WRXDExBTOa2oLODRFq-RIZXGsI-c,35013
|
411
415
|
sglang/srt/layers/quantization/compressed_tensors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
412
416
|
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=DsPCI82A4rqI6mjKo-WorIKUweppFV7-16Qku-zW9_g,25897
|
413
|
-
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=
|
417
|
+
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=dceX06gCeqiY2Ydw3o7s_YioVt4twEMnbnr5dM1Ff1M,24343
|
414
418
|
sglang/srt/layers/quantization/compressed_tensors/utils.py,sha256=mnUmKWFQUnY8bVoFHUuNVwqsfS-cefeR-ofyaihCXcY,7621
|
415
419
|
sglang/srt/layers/quantization/compressed_tensors/schemes/__init__.py,sha256=HWMTnmrj-mUCRXgcOwnnXLrvrAE-ONdPTSzSImjHCMA,347
|
416
420
|
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_scheme.py,sha256=tdKJC8c3SX8T3z8JL-1YCsg4ftcv55Wxt0vZrYftpX8,1635
|
@@ -569,14 +573,14 @@ sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a
|
|
569
573
|
"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
|
570
574
|
"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
|
571
575
|
sglang/srt/layers/quantization/deep_gemm_wrapper/__init__.py,sha256=esJMd0Yuj68t6QYOpmIFuiWP2J2dxTMC4bRBNH0Xk6I,26
|
572
|
-
sglang/srt/layers/quantization/deep_gemm_wrapper/compile_utils.py,sha256=
|
576
|
+
sglang/srt/layers/quantization/deep_gemm_wrapper/compile_utils.py,sha256=FaHKgY9y1oi3zlCmkorSK4r4dhPy7NiviVIhtabvzjM,11536
|
573
577
|
sglang/srt/layers/quantization/deep_gemm_wrapper/configurer.py,sha256=zkMzBAqQTSwG9TlGPuwmHV1yN0v5Y_5Hgyo0CKQDVNU,741
|
574
578
|
sglang/srt/layers/quantization/deep_gemm_wrapper/entrypoint.py,sha256=On_uvP1ThxgQFAzOWKoteM0vIOHqmtI127A4qbPtmbc,3264
|
575
579
|
sglang/srt/lora/layers.py,sha256=FaTYf0w3zbKKKAT1nJdAzGRtxvySB6FpNKXvEi5RHH0,12958
|
576
580
|
sglang/srt/lora/lora.py,sha256=sRoU7BdlTV4cnBv1-u3QjisVR9XQrrqAsmhxdNtd8dU,9433
|
577
581
|
sglang/srt/lora/lora_config.py,sha256=qDgMTx_69jyJUl29O5FxLzYa0BMhqYVXWXfyyVOvGm0,1684
|
578
582
|
sglang/srt/lora/lora_manager.py,sha256=ABuqg1IhY9LNXgJd_v3Yx_VHS52t4SapHmhLiH1thdc,20147
|
579
|
-
sglang/srt/lora/lora_registry.py,sha256=
|
583
|
+
sglang/srt/lora/lora_registry.py,sha256=iBPZFJq2JTa0p1pA_TJ5alU3V95NmazKai4GiDmVXnA,7865
|
580
584
|
sglang/srt/lora/mem_pool.py,sha256=UYvTgEPvBoVsaX5rEnipebOAlo75pML6_NO9GfCFzl8,12148
|
581
585
|
sglang/srt/lora/utils.py,sha256=5G0KiDEn-Zg4OgIOjWnU2ZdQGNstZswoAog9xU4GVDA,5660
|
582
586
|
sglang/srt/lora/backend/base_backend.py,sha256=EIz8I-GIrdmK4fISw3ENhbJVVITaxKfyLxHXGPU4fPs,5044
|
@@ -587,52 +591,58 @@ sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=QURCYxHNR8Ls4SQtt3dvdgjvdDVh
|
|
587
591
|
sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=iz5scFNo2YFTeiM8beWg9Z1oZI-6AM_T1wBMCQ6qp2Q,6485
|
588
592
|
sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=Ksova04wgeGsFqGOXWqJtMYaHgyUYcx8VU42BZQOkVA,5129
|
589
593
|
sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=VqCAFvUtq_l-0RGIkx3W_fzD55QcW2FAcqpvSHOBFp4,5115
|
590
|
-
sglang/srt/managers/cache_controller.py,sha256=
|
594
|
+
sglang/srt/managers/cache_controller.py,sha256=sqKPWX0X-q6KjkfvT3UXANGxpG8JvDfm4kLPTC5YvMo,29328
|
591
595
|
sglang/srt/managers/configure_logging.py,sha256=8sNXZ2z9pBWOwn-X3wyz013Ob8Nbm1zDxRkxoZjH-l4,1633
|
592
|
-
sglang/srt/managers/data_parallel_controller.py,sha256=
|
596
|
+
sglang/srt/managers/data_parallel_controller.py,sha256=VvEkidmYCGAFTv6upEHf_tyTvkM8xLhCyOnLJGWL6BQ,14592
|
593
597
|
sglang/srt/managers/detokenizer_manager.py,sha256=SpLxTsSPKBZfD-ZMhJ5zpPPGuUb8PmcYgFSL9CsurU4,10696
|
594
|
-
sglang/srt/managers/io_struct.py,sha256=
|
598
|
+
sglang/srt/managers/io_struct.py,sha256=hNYPIMEPtDRSHIkxpwD00capl4HAM-64kyIXOUgfk8k,38103
|
595
599
|
sglang/srt/managers/mm_utils.py,sha256=VrJiv0k_zDa83BsfVv8rmckgXKsvF0Wz88bCAb6R1as,29601
|
596
600
|
sglang/srt/managers/multimodal_processor.py,sha256=mLnGcUHIvjruwoVhqiz4j7RMYMsZnvQMH8JnKCITDlY,1805
|
597
|
-
sglang/srt/managers/schedule_batch.py,sha256=
|
598
|
-
sglang/srt/managers/schedule_policy.py,sha256=
|
599
|
-
sglang/srt/managers/scheduler.py,sha256=
|
601
|
+
sglang/srt/managers/schedule_batch.py,sha256=pHjIJAdfxNxfMB3Be4AV2yq-MaSN-CkA2yhREeHaxH4,76355
|
602
|
+
sglang/srt/managers/schedule_policy.py,sha256=l_2C8z52pSi-FOQpF_aR4joj9UL5wP9A0oJlIzSq6bM,22449
|
603
|
+
sglang/srt/managers/scheduler.py,sha256=oin_9RdC7jiJ-s8XkvuH5HZRSNQojjIsK70t4TyIaFc,106152
|
600
604
|
sglang/srt/managers/scheduler_input_blocker.py,sha256=zP8xU_UmU2H0AB6sEqvivDcDB1QDgTauNDYDIJ7Nez4,3683
|
601
605
|
sglang/srt/managers/scheduler_metrics_mixin.py,sha256=-1IOE30eOoTvvpz-Xg9J4UcamwAJle2-odFJ9vrinLE,9341
|
602
606
|
sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=Ei-9lyfrSQAlL9_ULuJLAhA7qpvUzjsBDzBBTDdnt4Q,31394
|
603
607
|
sglang/srt/managers/scheduler_profiler_mixin.py,sha256=GnR1SALuj_JbsENfmpiVgVcJWtGQn8hJ8TERkLB7h1c,10701
|
604
608
|
sglang/srt/managers/scheduler_update_weights_mixin.py,sha256=cQYdfHSgqLR7FeWMZl1V_lXMSKw99WGUa9gdiYUMXLU,5334
|
605
609
|
sglang/srt/managers/session_controller.py,sha256=dzlMNZlo20FTSl64QqK7y7pElsdCy8ICOWWBPTBVwgs,6040
|
606
|
-
sglang/srt/managers/template_manager.py,sha256=
|
607
|
-
sglang/srt/managers/tokenizer_manager.py,sha256=
|
610
|
+
sglang/srt/managers/template_manager.py,sha256=U79zdED8XSO5Ee9xF13QuPrjI3dVhzAuekRK94DDy_s,10682
|
611
|
+
sglang/srt/managers/tokenizer_manager.py,sha256=WSMH39L3OxKxbV_WZnBE2e5uUektc21BOzPCX5bBoz8,81929
|
608
612
|
sglang/srt/managers/tp_worker.py,sha256=Y8GEgEI9Y4ayw4K71yRHDzeVX43JjTBkf3vFf8RfS9w,11764
|
609
613
|
sglang/srt/managers/tp_worker_overlap_thread.py,sha256=3KQQ2YkyswgtouByhfPPq5mCISsMstn9CEa1i4qv-Xg,10998
|
610
|
-
sglang/srt/managers/utils.py,sha256=
|
614
|
+
sglang/srt/managers/utils.py,sha256=RogDdqb2bcxpNk5TbxKyaXFpgFz8t30IOnDp4s-IilM,2776
|
611
615
|
sglang/srt/mem_cache/allocator.py,sha256=id4riucsG9d-P1ikj3-DLJbzhM8-MuJfvDmB8H4g9-o,23364
|
612
616
|
sglang/srt/mem_cache/base_prefix_cache.py,sha256=hLS2ncTMAz7Kpdk5pNwn5c6g8b61_K9OCBL4Vj1xsYc,2801
|
613
617
|
sglang/srt/mem_cache/chunk_cache.py,sha256=FU_fq8e_dodgU8EkRqE4Jr0sKwCXQnKmapO88k6_JfU,3207
|
614
618
|
sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
|
615
|
-
sglang/srt/mem_cache/hicache_storage.py,sha256=
|
616
|
-
sglang/srt/mem_cache/hiradix_cache.py,sha256=
|
617
|
-
sglang/srt/mem_cache/memory_pool.py,sha256=
|
618
|
-
sglang/srt/mem_cache/memory_pool_host.py,sha256=
|
619
|
+
sglang/srt/mem_cache/hicache_storage.py,sha256=USRuV4bvePMxkKr1ypiDnZZSUBbE8Arm5PnZklsbrkc,6385
|
620
|
+
sglang/srt/mem_cache/hiradix_cache.py,sha256=QbKT2NKLuVe-1AqEc_njtZHXrq8d6WiNNbLa6Qio9qI,25818
|
621
|
+
sglang/srt/mem_cache/memory_pool.py,sha256=i-QhFrYuaAFcMOG7QH6SJyJ8hFzVRS5fjxwKmae8f94,37064
|
622
|
+
sglang/srt/mem_cache/memory_pool_host.py,sha256=M5dTu4BiEEiTlD7vtctd_LQrCODS1iradGaFtHaRkXM,25385
|
619
623
|
sglang/srt/mem_cache/multimodal_cache.py,sha256=wZl2KeEl3xeoEsYdH33UoM-FO8kqfLo_XUgereJVvoM,1348
|
620
624
|
sglang/srt/mem_cache/radix_cache.py,sha256=kp4JiVxh3eS4MsFxGKIYjYAb14B777qc1bHrABPqBMc,19101
|
625
|
+
sglang/srt/mem_cache/radix_cache_cpp.py,sha256=1V-XxYC2jxQYUMJBLrlcSMcqhCBo8TC7wRTTVbOaMkY,9269
|
621
626
|
sglang/srt/mem_cache/swa_radix_cache.py,sha256=4qt_gS7GuHiekd6VvKkx54dB2GHJzNA_kj71Np6jTDA,40300
|
622
|
-
sglang/srt/mem_cache/
|
623
|
-
sglang/srt/mem_cache/
|
624
|
-
sglang/srt/mem_cache/storage/hf3fs/
|
627
|
+
sglang/srt/mem_cache/cpp_radix_tree/radix_tree.py,sha256=tQZpz-H6HxNAQe9mYqt6aX8mfDZyu_fbJTcCNK-Ns1M,7179
|
628
|
+
sglang/srt/mem_cache/storage/hf3fs/client_hf3fs.py,sha256=k9ZtqKfWdJh19jLyRu0iXJEciXXuqFcEuMXej3hJhmg,5178
|
629
|
+
sglang/srt/mem_cache/storage/hf3fs/hf3fs_utils.cpp,sha256=dDYPtxmL6OFhEyv_GuTgzkTDpBjcD0EK8kphnn7eaGc,1133
|
625
630
|
sglang/srt/mem_cache/storage/hf3fs/storage_hf3fs.py,sha256=H0X4v5AWs4uTHoBGgb9BgfKf3WZ3reZKpLNxmL1dTbI,9230
|
626
631
|
sglang/srt/mem_cache/storage/hf3fs/test_hf3fs_utils.py,sha256=g2h0Woa2CK39pRTigHCVR_hq095dTaC70lCIAIs9THM,1089
|
632
|
+
sglang/srt/mem_cache/storage/mooncake_store/mooncake_store.py,sha256=vFDsTE1a-6FP_AidlGKrlX9-oZvdxh8rnQiFDd-UcE8,9748
|
633
|
+
sglang/srt/mem_cache/storage/mooncake_store/unit_test.py,sha256=jMdJgU41VscKS-mn_AVZE8LQbb_3MWQ57i9V3OCWLI4,1007
|
634
|
+
sglang/srt/mem_cache/storage/nixl/hicache_nixl.py,sha256=GU2J2335razBIIEqV2L44CR8N-QYKmhbHEDGa61EFEE,6170
|
635
|
+
sglang/srt/mem_cache/storage/nixl/nixl_utils.py,sha256=i2C70Ct2xdZtR_6OSL7uQk7ped1TieSpdUKo8yXdc08,8672
|
636
|
+
sglang/srt/mem_cache/storage/nixl/test_hicache_nixl_storage.py,sha256=TBtN12muLFXZ296mK2d3TSTyj8qhMo5kAYWLpSjLBhY,8017
|
627
637
|
sglang/srt/metrics/collector.py,sha256=FH6Wil2lMK7d7q74CfxfdULCaHKDXeAoo8UBZqqdepM,20289
|
628
638
|
sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
|
629
|
-
sglang/srt/model_executor/cuda_graph_runner.py,sha256=
|
630
|
-
sglang/srt/model_executor/forward_batch_info.py,sha256=
|
631
|
-
sglang/srt/model_executor/model_runner.py,sha256=
|
639
|
+
sglang/srt/model_executor/cuda_graph_runner.py,sha256=DAJwRTRqxm-D2Mp-ZhshL_WjMQTg0IKI5f1Jht4FO-o,33281
|
640
|
+
sglang/srt/model_executor/forward_batch_info.py,sha256=Drc3TsN5XJNkaRiJXHmCTwIazXsIEfmNi6TdQfBpJG4,38831
|
641
|
+
sglang/srt/model_executor/model_runner.py,sha256=4tWBilTr2ONfwsLuQZEkXK9qtxHyUcVxaJOwxmM6IYw,73387
|
632
642
|
sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
|
633
643
|
sglang/srt/model_loader/loader.py,sha256=wxSUYrqv23w9yHYy9Lm0ozWggs9pf6Co4yqe1UO8PZk,59799
|
634
644
|
sglang/srt/model_loader/utils.py,sha256=svSrPQxViBz_-0_sByu4coCYrzJ51-4FCi_bFqa6rDo,4474
|
635
|
-
sglang/srt/model_loader/weight_utils.py,sha256=
|
645
|
+
sglang/srt/model_loader/weight_utils.py,sha256=2UxbNYzoZxIIrDhsEOk6M-ofFg5s1UMtV1j23XRbnq4,38129
|
636
646
|
sglang/srt/models/arcee.py,sha256=Q6fujRESEnljzs8ls3XGsPtWxzfKfUa0QAvAekYzeBY,18981
|
637
647
|
sglang/srt/models/baichuan.py,sha256=HbvlErnkCSK4pRQYCSDxMcrn-1DQyfiNoeDcnRrJas8,15807
|
638
648
|
sglang/srt/models/bert.py,sha256=ODJe8YfNRP-hHsomFWk4_QpcuiSsNfjzGf256EDS0Pc,15802
|
@@ -643,7 +653,7 @@ sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,155
|
|
643
653
|
sglang/srt/models/deepseek.py,sha256=sqobTr9a9c5pNLNJZE-WrJIE_qBq0lC5gCGv9NpGU_Q,17364
|
644
654
|
sglang/srt/models/deepseek_janus_pro.py,sha256=td8xGs6ARfJ8AQCYwUhMOZoWigrAs7m3trF5-kXCqik,70418
|
645
655
|
sglang/srt/models/deepseek_nextn.py,sha256=47fehxRdiOizr0rdLg5f1fzQEx6gGAOcDcWKtblloyk,5928
|
646
|
-
sglang/srt/models/deepseek_v2.py,sha256=
|
656
|
+
sglang/srt/models/deepseek_v2.py,sha256=e28nSn6NJq9YlsbJWPqotIKBJr66Z9XyBPBk44K0azw,104084
|
647
657
|
sglang/srt/models/deepseek_vl2.py,sha256=7X3gI4DE7guUtNJvMLf06jO8MUHKa2Aan8evZg5hsXc,13061
|
648
658
|
sglang/srt/models/exaone.py,sha256=TpO-rtCpEZ8Ua7hGFnS8l2oAYhY0Pij50grc9WQ2mvc,13576
|
649
659
|
sglang/srt/models/gemma.py,sha256=2atKwFN5wCsJCmBRmURX9vaCcFjxlFirL7xsnG5KPlI,14180
|
@@ -655,13 +665,13 @@ sglang/srt/models/gemma3n_audio.py,sha256=isgKfjA5UieYawxU6medL2ssXlzYPqAbagDBnL
|
|
655
665
|
sglang/srt/models/gemma3n_causal.py,sha256=kBDlpSWAg6ZDmUPe08Q7sj5CbdVD_0h07zwJw8sKu7I,36335
|
656
666
|
sglang/srt/models/gemma3n_mm.py,sha256=pL_fvnkz9fKaC2Afms5KG3Me3u5sp1C2-mAUhqPGrQ4,18701
|
657
667
|
sglang/srt/models/glm4.py,sha256=2VQzUqFkQTy_2nfkxP9SF6_9kKLTZUExGRjge7r99Es,11265
|
658
|
-
sglang/srt/models/glm4_moe.py,sha256=
|
668
|
+
sglang/srt/models/glm4_moe.py,sha256=CXXPvglPPWWSmeEkf2x5iezp_-tBWY_dcDKqG6v6zzg,42890
|
659
669
|
sglang/srt/models/glm4_moe_nextn.py,sha256=cWB6o3w7xBvgi2ouGKsqZ3iovKH65QiRQul42wcgzMk,5897
|
660
670
|
sglang/srt/models/gpt2.py,sha256=kclhxEs8oJk1KCyhmAqo7rZqecVGGHYkc-a1WZi3aIk,9841
|
661
671
|
sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtYhl4,10307
|
662
672
|
sglang/srt/models/granite.py,sha256=5WOJyNYAlt5RNHSexNfPNihhSxIMd7wPzju1cTixKig,20852
|
663
673
|
sglang/srt/models/granitemoe.py,sha256=Yr1pEOS_8QopQKBIiDpFpMXZ0Wkc1REyPorM3eDTjNE,13808
|
664
|
-
sglang/srt/models/grok.py,sha256=
|
674
|
+
sglang/srt/models/grok.py,sha256=M23KdbuB3rrJoOG2LkS6EvVts1p0fMfmYAzcbjXQy-Q,28243
|
665
675
|
sglang/srt/models/hunyuan.py,sha256=RSZErx-Swt65kmfvSXJQJJ6KhjucuNY4UUVLWC8hWaA,31102
|
666
676
|
sglang/srt/models/idefics2.py,sha256=U3khd3hbdawJeRNXsxmaKHdssOCT5TPOZ1D-2_zHoQo,12079
|
667
677
|
sglang/srt/models/internlm2.py,sha256=F_iNY1gUqzAjAuUatcE47gnrcoTh5_08PY2Rw9tKr9M,13150
|
@@ -671,7 +681,7 @@ sglang/srt/models/internvl.py,sha256=o6XSAvnMYK5tTPbln9RI9BfRfAQZBE8_xmGPffrznnQ
|
|
671
681
|
sglang/srt/models/kimi_vl.py,sha256=XulRiufQzOSWZe8-bUl2OEdznvsVU9OBsEYipaX3YkM,12877
|
672
682
|
sglang/srt/models/kimi_vl_moonvit.py,sha256=M5L7AJOJ2Zh2pqLZAq8aRqhbTSlipr8XOoD3ix6o2sU,23908
|
673
683
|
sglang/srt/models/llama.py,sha256=Nsxv2s1sOmLnZsTHkwbTh53cUFEQ2fnKDQi5BFbFcsk,28994
|
674
|
-
sglang/srt/models/llama4.py,sha256=
|
684
|
+
sglang/srt/models/llama4.py,sha256=frV2ZWeupnHMnjsurP2HCARluhjncvL_ydRXwg447nE,19379
|
675
685
|
sglang/srt/models/llama_classification.py,sha256=4QWTFaUZIFKYZvEzs8bx8VkOZNIwdYCLrnwrdAw4QK0,3108
|
676
686
|
sglang/srt/models/llama_eagle.py,sha256=Ubh_jmtfiOnriwvHgQT0ZGID7JvYvdSi_QGHOIkEgPE,5028
|
677
687
|
sglang/srt/models/llama_eagle3.py,sha256=715soylDE3kOMWskbPy_70BFxaSc5ptaHCS1v3-ViA0,8579
|
@@ -686,10 +696,10 @@ sglang/srt/models/minicpm3.py,sha256=6-gfHSfXwyB5zw7AIj-c5TzjpEKR6CM0H67MrD8LVUE
|
|
686
696
|
sglang/srt/models/minicpmo.py,sha256=st8I13Ere5RkRHHQQN1LgZh1ybUA3_buxaaE4XlV8ag,76123
|
687
697
|
sglang/srt/models/minicpmv.py,sha256=g4aAOG8PVGkZqGmlneVaScAtNpSLqSVbZGRAikyrBy8,30394
|
688
698
|
sglang/srt/models/mistral.py,sha256=ni7PppA_5X4ivTnIps3yoEaMEL-G6Gaf9b5hGC1vU5Q,3475
|
689
|
-
sglang/srt/models/mixtral.py,sha256=
|
699
|
+
sglang/srt/models/mixtral.py,sha256=1zGxNHHAhhQkKJz3Kr7wFJ1-kMGVkwKYRyvVcJ0l4LQ,17266
|
690
700
|
sglang/srt/models/mixtral_quant.py,sha256=UUIq7lpUonD9IoGePyl59oJYEcVwT9wUXvtVp5tjQ9Y,15406
|
691
701
|
sglang/srt/models/mllama.py,sha256=OdX7brhyfGiMFPulxF77QOUi-Kcscg3Qo4ipyxzs9yA,39670
|
692
|
-
sglang/srt/models/mllama4.py,sha256=
|
702
|
+
sglang/srt/models/mllama4.py,sha256=eWjuYA7GZ6NrPFFr_qkA1JC5yRHrQ_oKnoelWTyDvp4,34745
|
693
703
|
sglang/srt/models/olmo.py,sha256=7-q_fA6XXdG7kPUjpUzYkzMUWJobuSjhqjYw9xSUs_c,12671
|
694
704
|
sglang/srt/models/olmo2.py,sha256=azmljhJF4ivcQfUtfsAUxq3ducE4tRKTL6iwe0IKYMg,14327
|
695
705
|
sglang/srt/models/olmoe.py,sha256=30HbQ35ELbJTwvD2fWezMoTt9zYYfsFDE5mxz4pDtgg,16129
|
@@ -707,16 +717,16 @@ sglang/srt/models/qwen2_5_vl.py,sha256=vOuIyK_NLuGgz9LlAdyM7cSQUU93VnBCYXUr_n4iO
|
|
707
717
|
sglang/srt/models/qwen2_audio.py,sha256=lVeYmC29eVZIR9I6ZjFe4xvC0wSXR0NtGCli6fSXtms,7870
|
708
718
|
sglang/srt/models/qwen2_classification.py,sha256=dGrMm4ebd30_lBhHOhaV57ig2iOTx3nqB4GEzsrRIM8,2747
|
709
719
|
sglang/srt/models/qwen2_eagle.py,sha256=6spFg6JYQGeUXLv2Q5eMkVzZvnqGyvW6SdVBtFMxlTM,4806
|
710
|
-
sglang/srt/models/qwen2_moe.py,sha256=
|
720
|
+
sglang/srt/models/qwen2_moe.py,sha256=Mgwe-yg22rkPOOic_ewkfb7C9DEQsx4irtUO9GXG4Vo,25664
|
711
721
|
sglang/srt/models/qwen2_rm.py,sha256=-mQXDEv11p-I1HXgYLTtY6ROem6UYorO958WsDrzsgs,2837
|
712
722
|
sglang/srt/models/qwen2_vl.py,sha256=ZRAL32Ymt4ZCBqB0JorXAm4ZBrECZ9EbaXKaTSikkQM,21366
|
713
723
|
sglang/srt/models/qwen3.py,sha256=COSC1TsCQNTq3E1QoVJYj1yTn-CFEVEX7keRwrrBXLs,20121
|
714
|
-
sglang/srt/models/qwen3_moe.py,sha256=
|
724
|
+
sglang/srt/models/qwen3_moe.py,sha256=C3zzu50iHEbTb5KUZp9I0672m6bruR0lShz4I5m19Lo,31434
|
715
725
|
sglang/srt/models/registry.py,sha256=bH9H0OiNV0Cn7FRBM30Oh6dxICTs9mJscdLtHYPJvxc,3783
|
716
726
|
sglang/srt/models/roberta.py,sha256=3k53V2Gbezk3jU-D03thXx8csGn5DxFK01ZQ9WFiWPI,9828
|
717
727
|
sglang/srt/models/siglip.py,sha256=DIVJRwdtpLD2QT1kVPIHw7Bn9BE40xJbkHujDNzkjgA,9892
|
718
728
|
sglang/srt/models/stablelm.py,sha256=0x_31uIr3WcWwecdPAI3ek9KkyKBJS7VwknTk2y0gjY,12281
|
719
|
-
sglang/srt/models/step3_vl.py,sha256=
|
729
|
+
sglang/srt/models/step3_vl.py,sha256=j7EdJN6zSsFkDAz9tRGTpKFMw6XJueJEIdFrTVZoALk,35138
|
720
730
|
sglang/srt/models/torch_native_llama.py,sha256=oTxxof5Wx2b2egIwWx56LeHjtPGQleQ2zM5jXHMiuOU,19293
|
721
731
|
sglang/srt/models/transformers.py,sha256=_CL0npi-_HvqzxkLgFhv2EE3bfGvCLXPYAtPFywZOr0,10260
|
722
732
|
sglang/srt/models/vila.py,sha256=3NiRHWOqXnCRjXmswAldsNtiE5zRILbScEsbhUzLZF4,9846
|
@@ -724,11 +734,11 @@ sglang/srt/models/xverse.py,sha256=DsNVI9JpzN4jj0Ry6aTrj7r-xq5YLOoDX2kH4YLJA-I,1
|
|
724
734
|
sglang/srt/models/xverse_moe.py,sha256=7KCM2-j12towDMNvXkuuYiBOmNauH6NG4Ip40x0khqA,16782
|
725
735
|
sglang/srt/models/yivl.py,sha256=oToK7-u5IGO7xwpJIQ7VtudlK6-zPqJX4bt6_wv0SH8,4850
|
726
736
|
sglang/srt/multimodal/mm_utils.py,sha256=6bV1_a7Iet9W0y3qdd0E-hFM78zalIzNZCYbLzJX2y8,12640
|
727
|
-
sglang/srt/multimodal/processors/base_processor.py,sha256=
|
737
|
+
sglang/srt/multimodal/processors/base_processor.py,sha256=ctdHYzzmplktfQYINHBN7GM3TLtiunM3n73yFulHAQI,24216
|
728
738
|
sglang/srt/multimodal/processors/clip.py,sha256=5nXZuQYbyf-5mlqPwjhQrlWqxvswImWZQXivChsiHfQ,1057
|
729
739
|
sglang/srt/multimodal/processors/deepseek_vl_v2.py,sha256=Pymag9f0U5L_bBOTFs2q94BbuPJ-VwqSixs3meO0nsM,2569
|
730
740
|
sglang/srt/multimodal/processors/gemma3.py,sha256=7aD6cR60iSC_MqjzdCDMAgtrb_cmy9YJyqLgVRClBT8,1986
|
731
|
-
sglang/srt/multimodal/processors/gemma3n.py,sha256=
|
741
|
+
sglang/srt/multimodal/processors/gemma3n.py,sha256=mtmvBasOMxb3Y2NLZR02nMYX30RLUXhDYJ5uVOElr10,2776
|
732
742
|
sglang/srt/multimodal/processors/internvl.py,sha256=p8dba1LBEceGObJvPdq3auDhlMKQIXyp_fXM5VMd4BI,9771
|
733
743
|
sglang/srt/multimodal/processors/janus_pro.py,sha256=_4ltWmdtFjbh1VCMRbcgxW61eWIABIzHG-YgwHWaYq8,1471
|
734
744
|
sglang/srt/multimodal/processors/kimi_vl.py,sha256=5i8KmGxnnDtQc7XqUBNyanAfGxhxj8iRFBFONHmpiFg,1565
|
@@ -752,16 +762,16 @@ sglang/srt/sampling/penaltylib/orchestrator.py,sha256=XM-Lm6u7gYPtMZrTIc0FR4QxNZ
|
|
752
762
|
sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=NRh10AJrrQlGJ6S-enGdRefrTrWpyqrSm-aNnyqQNQQ,2119
|
753
763
|
sglang/srt/speculative/build_eagle_tree.py,sha256=O9LJNaBflgJdWT94D-rGH1gJFJ18nst2oOD8HnA2mZ4,12859
|
754
764
|
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=kVqMGj25jxtJlP-nTSlq_QJgoabFqKOU7-0WeUFhvw8,14509
|
755
|
-
sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py,sha256=
|
765
|
+
sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py,sha256=lrh3ZUNX2KAPtSVuTguxHKCJk3WhbaaW5ADEudC_PyM,16069
|
756
766
|
sglang/srt/speculative/eagle_utils.py,sha256=x5MKA2GkecNAXE58KKhPJRy5KxCLhRkD-C_kVSZIyzc,46433
|
757
767
|
sglang/srt/speculative/eagle_worker.py,sha256=AWMaTVgNgKmCecLsQRMLR5jIW6j69ogAQewYHrPK4d0,38717
|
758
768
|
sglang/srt/speculative/spec_info.py,sha256=rhaKG0TzyF9XZYHEWp1jccwTBohSNsUDvxHFtAoOl18,709
|
759
|
-
sglang/srt/weight_sync/utils.py,sha256=
|
769
|
+
sglang/srt/weight_sync/utils.py,sha256=43zl3u8Svzj49p7Sk78xp7SWW83mlr3EH-OZFGUGy5w,4157
|
760
770
|
sglang/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
761
771
|
sglang/test/few_shot_gsm8k.py,sha256=qnEejCyPHGEMBEiNc7T4-t9lvoo2DNN72RXw-ei_TnY,4300
|
762
772
|
sglang/test/few_shot_gsm8k_engine.py,sha256=Yf1Dg0Tk-tY4v1oj_UTxJ7950CWn6kRMbQKDjT2Orgo,3902
|
763
773
|
sglang/test/run_eval.py,sha256=K4Z3OdMPvLeuGJhWWTBD0F-MOUFeBYLvsPnDQ44bd18,4036
|
764
|
-
sglang/test/runners.py,sha256=
|
774
|
+
sglang/test/runners.py,sha256=5kCwKrww_TQQSoSV3AMnWPl_qG_l9tH-OKQEgxwAxJQ,33731
|
765
775
|
sglang/test/send_one.py,sha256=XNZdfZs21tXDBwrDj2pq46WQd5smiKoYETY8U09c1Z8,5079
|
766
776
|
sglang/test/simple_eval_common.py,sha256=OW4Z7O33QqDySRFuHIC4Ouidfup4eaOEuR9i7xJzkDY,12379
|
767
777
|
sglang/test/simple_eval_gpqa.py,sha256=8Xt9Bw05c7SZTYrCZgB68OZUqUbLo69ywiyx0bTvSUk,3220
|
@@ -783,13 +793,14 @@ sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg
|
|
783
793
|
sglang/test/test_marlin_moe.py,sha256=PITMGcGF4ci2csXJejdLSNuxpURnFmcflsipAZsGakU,7991
|
784
794
|
sglang/test/test_marlin_utils.py,sha256=gFMJwxbfEEJDeHI-HU5FDF9FFqqcM-yahK0TWyVIVgQ,5171
|
785
795
|
sglang/test/test_programs.py,sha256=DJIbTChloPwaDiEWHQoSTuKv3BGx0-K8Y5e1mDDTACM,18910
|
786
|
-
sglang/test/test_utils.py,sha256=
|
796
|
+
sglang/test/test_utils.py,sha256=cCvqi0eLPaOK41a351iji4g1cVmm6VRpkpuGSdrZF28,41131
|
787
797
|
sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
788
798
|
sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
|
789
799
|
sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
|
790
800
|
sglang/test/attention/test_prefix_chunk_info.py,sha256=hpoDe2wfSa6RlUbfyri_c0iyBTb35UXGL9I2Xh6jamM,7772
|
791
|
-
sglang
|
792
|
-
sglang-0.4.10.dist-info/
|
793
|
-
sglang-0.4.10.dist-info/
|
794
|
-
sglang-0.4.10.dist-info/
|
795
|
-
sglang-0.4.10.dist-info/
|
801
|
+
sglang/test/attention/test_trtllm_mla_backend.py,sha256=D7KOJbsfGRSdO5dMlGokBpiC6Sh42g51HlVkgxSK-yI,36251
|
802
|
+
sglang-0.4.10.post2.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
|
803
|
+
sglang-0.4.10.post2.dist-info/METADATA,sha256=j--rk52lslAMmr3aXQyGrZ9W8nJa9RurXTh8ENc_twQ,27433
|
804
|
+
sglang-0.4.10.post2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
|
805
|
+
sglang-0.4.10.post2.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
|
806
|
+
sglang-0.4.10.post2.dist-info/RECORD,,
|
File without changes
|
File without changes
|
File without changes
|
File without changes
|
File without changes
|