sglang 0.5.2rc2__py3-none-any.whl → 0.5.3__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- sglang/bench_one_batch.py +7 -9
- sglang/bench_one_batch_server.py +330 -31
- sglang/bench_serving.py +267 -32
- sglang/global_config.py +2 -2
- sglang/lang/backend/runtime_endpoint.py +1 -1
- sglang/launch_server.py +14 -0
- sglang/profiler.py +2 -2
- sglang/srt/batch_invariant_ops/__init__.py +27 -0
- sglang/srt/batch_invariant_ops/batch_invariant_ops.py +549 -0
- sglang/srt/configs/__init__.py +8 -0
- sglang/srt/configs/device_config.py +3 -1
- sglang/srt/configs/dots_ocr.py +64 -0
- sglang/srt/configs/dots_vlm.py +139 -0
- sglang/srt/configs/falcon_h1.py +360 -0
- sglang/srt/configs/load_config.py +9 -0
- sglang/srt/configs/model_config.py +181 -82
- sglang/srt/configs/qwen3_next.py +326 -0
- sglang/srt/configs/qwen3_vl.py +586 -0
- sglang/srt/connector/__init__.py +8 -1
- sglang/srt/connector/remote_instance.py +82 -0
- sglang/srt/constrained/base_grammar_backend.py +49 -12
- sglang/srt/constrained/llguidance_backend.py +0 -1
- sglang/srt/constrained/outlines_backend.py +0 -1
- sglang/srt/constrained/outlines_jump_forward.py +1 -1
- sglang/srt/constrained/xgrammar_backend.py +30 -9
- sglang/srt/custom_op.py +11 -1
- sglang/srt/debug_utils/dump_comparator.py +81 -44
- sglang/srt/debug_utils/dump_loader.py +97 -0
- sglang/srt/debug_utils/dumper.py +21 -6
- sglang/srt/debug_utils/text_comparator.py +73 -11
- sglang/srt/disaggregation/ascend/conn.py +2 -2
- sglang/srt/disaggregation/ascend/transfer_engine.py +47 -9
- sglang/srt/disaggregation/base/conn.py +1 -1
- sglang/srt/disaggregation/common/conn.py +279 -108
- sglang/srt/disaggregation/decode.py +71 -19
- sglang/srt/disaggregation/decode_kvcache_offload_manager.py +185 -0
- sglang/srt/disaggregation/decode_schedule_batch_mixin.py +29 -17
- sglang/srt/disaggregation/fake/conn.py +1 -1
- sglang/srt/disaggregation/mini_lb.py +6 -445
- sglang/srt/disaggregation/mooncake/conn.py +55 -537
- sglang/srt/disaggregation/nixl/conn.py +326 -53
- sglang/srt/disaggregation/prefill.py +36 -17
- sglang/srt/disaggregation/utils.py +40 -54
- sglang/srt/distributed/device_communicators/all_reduce_utils.py +16 -0
- sglang/srt/distributed/device_communicators/shm_broadcast.py +4 -2
- sglang/srt/distributed/device_communicators/symm_mem.py +164 -0
- sglang/srt/distributed/parallel_state.py +156 -80
- sglang/srt/entrypoints/engine.py +59 -18
- sglang/srt/entrypoints/grpc_request_manager.py +855 -0
- sglang/srt/entrypoints/grpc_server.py +810 -0
- sglang/srt/entrypoints/http_server.py +130 -59
- sglang/srt/entrypoints/openai/protocol.py +112 -4
- sglang/srt/entrypoints/openai/serving_base.py +65 -3
- sglang/srt/entrypoints/openai/serving_chat.py +204 -55
- sglang/srt/entrypoints/openai/serving_completions.py +14 -3
- sglang/srt/entrypoints/openai/serving_embedding.py +9 -3
- sglang/srt/entrypoints/openai/serving_rerank.py +3 -1
- sglang/srt/entrypoints/openai/serving_responses.py +48 -3
- sglang/srt/entrypoints/openai/serving_score.py +1 -0
- sglang/srt/environ.py +285 -0
- sglang/srt/eplb/eplb_manager.py +2 -2
- sglang/srt/eplb/expert_distribution.py +26 -13
- sglang/srt/eplb/expert_location.py +38 -8
- sglang/srt/eplb/expert_location_updater.py +1 -1
- sglang/srt/function_call/base_format_detector.py +3 -6
- sglang/srt/function_call/ebnf_composer.py +11 -9
- sglang/srt/function_call/function_call_parser.py +9 -2
- sglang/srt/function_call/glm4_moe_detector.py +4 -4
- sglang/srt/function_call/gpt_oss_detector.py +23 -0
- sglang/srt/function_call/json_array_parser.py +63 -0
- sglang/srt/function_call/kimik2_detector.py +17 -4
- sglang/srt/function_call/qwen3_coder_detector.py +1 -1
- sglang/srt/function_call/utils.py +96 -5
- sglang/srt/grpc/__init__.py +1 -0
- sglang/srt/grpc/compile_proto.py +245 -0
- sglang/srt/grpc/sglang_scheduler_pb2.py +111 -0
- sglang/srt/grpc/sglang_scheduler_pb2.pyi +434 -0
- sglang/srt/grpc/sglang_scheduler_pb2_grpc.py +239 -0
- sglang/srt/layers/activation.py +143 -9
- sglang/srt/layers/attention/aiter_backend.py +14 -15
- sglang/srt/layers/attention/ascend_backend.py +115 -9
- sglang/srt/layers/attention/attention_registry.py +206 -0
- sglang/srt/layers/attention/base_attn_backend.py +12 -3
- sglang/srt/layers/attention/cutlass_mla_backend.py +3 -3
- sglang/srt/layers/attention/dual_chunk_flashattention_backend.py +1 -1
- sglang/srt/layers/attention/fla/chunk.py +242 -0
- sglang/srt/layers/attention/fla/chunk_delta_h.py +314 -0
- sglang/srt/layers/attention/fla/chunk_o.py +178 -0
- sglang/srt/layers/attention/fla/chunk_scaled_dot_kkt.py +151 -0
- sglang/srt/layers/attention/fla/cumsum.py +300 -0
- sglang/srt/layers/attention/fla/fused_recurrent.py +640 -0
- sglang/srt/layers/attention/fla/fused_sigmoid_gating_recurrent.py +232 -0
- sglang/srt/layers/attention/fla/index.py +37 -0
- sglang/srt/layers/attention/fla/l2norm.py +150 -0
- sglang/srt/layers/attention/fla/layernorm_gated.py +326 -0
- sglang/srt/layers/attention/fla/op.py +66 -0
- sglang/srt/layers/attention/fla/solve_tril.py +465 -0
- sglang/srt/layers/attention/fla/utils.py +331 -0
- sglang/srt/layers/attention/fla/wy_fast.py +158 -0
- sglang/srt/layers/attention/flashattention_backend.py +41 -8
- sglang/srt/layers/attention/flashinfer_backend.py +118 -198
- sglang/srt/layers/attention/flashinfer_mla_backend.py +27 -27
- sglang/srt/layers/attention/flashmla_backend.py +7 -5
- sglang/srt/layers/attention/hybrid_attn_backend.py +68 -53
- sglang/srt/layers/attention/hybrid_linear_attn_backend.py +602 -0
- sglang/srt/layers/attention/intel_amx_backend.py +3 -0
- sglang/srt/layers/attention/mamba/causal_conv1d.py +129 -0
- sglang/srt/layers/attention/mamba/causal_conv1d_triton.py +969 -0
- sglang/srt/layers/attention/mamba/mamba.py +629 -0
- sglang/srt/layers/attention/mamba/mamba_utils.py +81 -0
- sglang/srt/layers/attention/mamba/ops/__init__.py +2 -0
- sglang/srt/layers/attention/mamba/ops/layernorm_gated.py +172 -0
- sglang/srt/layers/attention/mamba/ops/mamba_ssm.py +442 -0
- sglang/srt/layers/attention/mamba/ops/ssd_bmm.py +264 -0
- sglang/srt/layers/attention/mamba/ops/ssd_chunk_scan.py +622 -0
- sglang/srt/layers/attention/mamba/ops/ssd_chunk_state.py +757 -0
- sglang/srt/layers/attention/mamba/ops/ssd_combined.py +262 -0
- sglang/srt/layers/attention/mamba/ops/ssd_state_passing.py +275 -0
- sglang/srt/layers/attention/npu_ops/mla_preprocess.py +393 -0
- sglang/srt/layers/attention/nsa/dequant_k_cache.py +163 -0
- sglang/srt/layers/attention/nsa/index_buf_accessor.py +354 -0
- sglang/srt/layers/attention/nsa/nsa_indexer.py +761 -0
- sglang/srt/layers/attention/nsa/quant_k_cache.py +255 -0
- sglang/srt/layers/attention/nsa/tilelang_kernel.py +785 -0
- sglang/srt/layers/attention/nsa/transform_index.py +144 -0
- sglang/srt/layers/attention/nsa/utils.py +24 -0
- sglang/srt/layers/attention/nsa_backend.py +887 -0
- sglang/srt/layers/attention/tbo_backend.py +6 -6
- sglang/srt/layers/attention/torch_flex_backend.py +325 -0
- sglang/srt/layers/attention/torch_native_backend.py +12 -6
- sglang/srt/layers/attention/triton_backend.py +57 -7
- sglang/srt/layers/attention/trtllm_mha_backend.py +5 -7
- sglang/srt/layers/attention/trtllm_mla_backend.py +276 -39
- sglang/srt/layers/attention/vision.py +58 -0
- sglang/srt/layers/attention/wave_backend.py +4 -4
- sglang/srt/layers/attention/wave_ops/decode_attention.py +2 -4
- sglang/srt/layers/attention/wave_ops/extend_attention.py +1 -3
- sglang/srt/layers/communicator.py +8 -0
- sglang/srt/layers/dp_attention.py +41 -2
- sglang/srt/layers/elementwise.py +3 -1
- sglang/srt/layers/layernorm.py +34 -15
- sglang/srt/layers/linear.py +55 -7
- sglang/srt/layers/logits_processor.py +44 -12
- sglang/srt/layers/moe/__init__.py +2 -1
- sglang/srt/layers/moe/cutlass_w4a8_moe.py +3 -3
- sglang/srt/layers/moe/ep_moe/kernels.py +2 -2
- sglang/srt/layers/moe/ep_moe/layer.py +256 -63
- sglang/srt/layers/moe/flashinfer_cutedsl_moe.py +183 -0
- sglang/srt/layers/moe/fused_moe_native.py +5 -3
- sglang/srt/layers/moe/fused_moe_triton/configs/{triton_3_4_0/E=128,N=768,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_3_1/E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json } +35 -35
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=352,device_name=NVIDIA_RTX_5880_Ada_Generation,dtype=fp8_w8a8.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=256,N=256,device_name=NVIDIA_H800,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=256,N=512,device_name=NVIDIA_H20.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H20-3e.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H200.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H800,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=256,device_name=NVIDIA_B200.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=256,device_name=NVIDIA_H20-3e.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=256,device_name=NVIDIA_H200.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=64,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=64,device_name=NVIDIA_H200.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +5 -2
- sglang/srt/layers/moe/fused_moe_triton/fused_moe_triton_config.py +7 -3
- sglang/srt/layers/moe/fused_moe_triton/fused_moe_triton_kernels.py +23 -20
- sglang/srt/layers/moe/fused_moe_triton/layer.py +71 -70
- sglang/srt/layers/moe/moe_runner/__init__.py +2 -1
- sglang/srt/layers/moe/moe_runner/base.py +274 -1
- sglang/srt/layers/moe/moe_runner/runner.py +80 -0
- sglang/srt/layers/moe/moe_runner/triton.py +448 -0
- sglang/srt/layers/moe/token_dispatcher/__init__.py +16 -4
- sglang/srt/layers/moe/token_dispatcher/{base_dispatcher.py → base.py} +67 -17
- sglang/srt/layers/moe/token_dispatcher/deepep.py +118 -56
- sglang/srt/layers/moe/token_dispatcher/standard.py +44 -2
- sglang/srt/layers/moe/topk.py +30 -9
- sglang/srt/layers/moe/utils.py +22 -6
- sglang/srt/layers/parameter.py +23 -6
- sglang/srt/layers/quantization/awq.py +19 -7
- sglang/srt/layers/quantization/base_config.py +11 -6
- sglang/srt/layers/quantization/blockwise_int8.py +38 -27
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py +1 -0
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py +50 -30
- sglang/srt/layers/quantization/compressed_tensors/schemes/__init__.py +2 -0
- sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_fp8.py +13 -1
- sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_int8.py +173 -0
- sglang/srt/layers/quantization/deep_gemm_wrapper/configurer.py +2 -10
- sglang/srt/layers/quantization/deep_gemm_wrapper/entrypoint.py +27 -0
- sglang/srt/layers/quantization/fp8.py +78 -49
- sglang/srt/layers/quantization/fp8_utils.py +51 -32
- sglang/srt/layers/quantization/gptq.py +25 -17
- sglang/srt/layers/quantization/modelopt_quant.py +190 -55
- sglang/srt/layers/quantization/moe_wna16.py +21 -18
- sglang/srt/layers/quantization/mxfp4.py +74 -42
- sglang/srt/layers/quantization/quark/quark_moe.py +48 -30
- sglang/srt/layers/quantization/unquant.py +135 -47
- sglang/srt/layers/quantization/w4afp8.py +26 -17
- sglang/srt/layers/quantization/w8a8_fp8.py +35 -20
- sglang/srt/layers/quantization/w8a8_int8.py +91 -41
- sglang/srt/layers/rotary_embedding.py +78 -31
- sglang/srt/layers/sampler.py +213 -21
- sglang/srt/layers/utils.py +23 -0
- sglang/srt/lora/backend/base_backend.py +50 -8
- sglang/srt/lora/backend/chunked_backend.py +348 -0
- sglang/srt/lora/backend/triton_backend.py +99 -5
- sglang/srt/lora/layers.py +32 -0
- sglang/srt/lora/lora.py +8 -3
- sglang/srt/lora/lora_manager.py +52 -118
- sglang/srt/lora/mem_pool.py +25 -11
- sglang/srt/lora/triton_ops/__init__.py +4 -0
- sglang/srt/lora/triton_ops/chunked_sgmv_expand.py +214 -0
- sglang/srt/lora/triton_ops/chunked_sgmv_shrink.py +174 -0
- sglang/srt/lora/utils.py +22 -11
- sglang/srt/managers/async_dynamic_batch_tokenizer.py +170 -0
- sglang/srt/managers/cache_controller.py +199 -301
- sglang/srt/managers/data_parallel_controller.py +115 -80
- sglang/srt/managers/detokenizer_manager.py +19 -15
- sglang/srt/managers/disagg_service.py +46 -0
- sglang/srt/managers/io_struct.py +340 -109
- sglang/srt/managers/mm_utils.py +44 -6
- sglang/srt/managers/multi_tokenizer_mixin.py +357 -407
- sglang/srt/managers/multimodal_processor.py +1 -2
- sglang/srt/managers/overlap_utils.py +53 -0
- sglang/srt/managers/schedule_batch.py +240 -138
- sglang/srt/managers/schedule_policy.py +144 -17
- sglang/srt/managers/scheduler.py +502 -209
- sglang/srt/managers/scheduler_input_blocker.py +1 -1
- sglang/srt/managers/scheduler_metrics_mixin.py +99 -126
- sglang/srt/managers/scheduler_output_processor_mixin.py +75 -22
- sglang/srt/managers/scheduler_profiler_mixin.py +6 -6
- sglang/srt/managers/scheduler_update_weights_mixin.py +7 -0
- sglang/srt/managers/tokenizer_communicator_mixin.py +675 -0
- sglang/srt/managers/tokenizer_manager.py +320 -632
- sglang/srt/managers/tp_worker.py +81 -22
- sglang/srt/managers/tp_worker_overlap_thread.py +71 -56
- sglang/srt/managers/utils.py +1 -45
- sglang/srt/mem_cache/allocator.py +14 -20
- sglang/srt/mem_cache/allocator_ascend.py +41 -27
- sglang/srt/mem_cache/base_prefix_cache.py +1 -1
- sglang/srt/mem_cache/chunk_cache.py +8 -1
- sglang/srt/mem_cache/evict_policy.py +23 -0
- sglang/srt/mem_cache/hicache_storage.py +43 -24
- sglang/srt/mem_cache/hiradix_cache.py +222 -75
- sglang/srt/mem_cache/memory_pool.py +535 -58
- sglang/srt/mem_cache/memory_pool_host.py +239 -228
- sglang/srt/mem_cache/radix_cache.py +222 -73
- sglang/srt/mem_cache/radix_cache_cpp.py +11 -8
- sglang/srt/mem_cache/storage/__init__.py +10 -0
- sglang/srt/mem_cache/storage/aibrix_kvcache/aibrix_kvcache_storage.py +151 -0
- sglang/srt/mem_cache/storage/aibrix_kvcache/unit_test.py +109 -0
- sglang/srt/mem_cache/storage/backend_factory.py +223 -0
- sglang/srt/mem_cache/storage/eic/eic_storage.py +778 -0
- sglang/srt/mem_cache/storage/eic/test_unit.py +115 -0
- sglang/srt/mem_cache/storage/hf3fs/hf3fs_client.py +164 -0
- sglang/srt/mem_cache/storage/hf3fs/{client_hf3fs.py → hf3fs_usrbio_client.py} +5 -1
- sglang/srt/mem_cache/storage/hf3fs/storage_hf3fs.py +259 -62
- sglang/srt/mem_cache/storage/lmcache/lmc_radix_cache.py +284 -0
- sglang/srt/mem_cache/storage/lmcache/unit_test.py +121 -0
- sglang/srt/mem_cache/storage/mooncake_store/mooncake_store.py +166 -17
- sglang/srt/mem_cache/swa_radix_cache.py +25 -36
- sglang/srt/metrics/collector.py +511 -132
- sglang/srt/metrics/func_timer.py +2 -7
- sglang/srt/metrics/startup_func_log_and_timer.py +150 -0
- sglang/srt/metrics/utils.py +8 -1
- sglang/srt/model_executor/cpu_graph_runner.py +640 -0
- sglang/srt/model_executor/cuda_graph_runner.py +52 -37
- sglang/srt/model_executor/forward_batch_info.py +82 -40
- sglang/srt/model_executor/model_runner.py +432 -157
- sglang/srt/model_executor/npu_graph_runner.py +12 -5
- sglang/srt/model_loader/__init__.py +9 -3
- sglang/srt/model_loader/loader.py +133 -5
- sglang/srt/model_loader/remote_instance_weight_loader_utils.py +69 -0
- sglang/srt/model_loader/weight_utils.py +158 -3
- sglang/srt/models/apertus.py +686 -0
- sglang/srt/models/bailing_moe.py +820 -217
- sglang/srt/models/bailing_moe_nextn.py +168 -0
- sglang/srt/models/deepseek_nextn.py +6 -1
- sglang/srt/models/deepseek_v2.py +607 -130
- sglang/srt/models/dots_ocr.py +173 -0
- sglang/srt/models/dots_vlm.py +174 -0
- sglang/srt/models/dots_vlm_vit.py +337 -0
- sglang/srt/models/ernie4.py +1 -1
- sglang/srt/models/falcon_h1.py +576 -0
- sglang/srt/models/gemma3_causal.py +0 -2
- sglang/srt/models/gemma3_mm.py +1 -1
- sglang/srt/models/gemma3n_mm.py +2 -2
- sglang/srt/models/glm4_moe.py +4 -4
- sglang/srt/models/glm4_moe_nextn.py +2 -2
- sglang/srt/models/glm4v.py +5 -3
- sglang/srt/models/glm4v_moe.py +4 -1
- sglang/srt/models/gpt_oss.py +8 -31
- sglang/srt/models/kimi_vl_moonvit.py +2 -2
- sglang/srt/models/llama.py +4 -0
- sglang/srt/models/llama4.py +9 -0
- sglang/srt/models/llama_eagle3.py +13 -0
- sglang/srt/models/longcat_flash.py +3 -3
- sglang/srt/models/longcat_flash_nextn.py +1 -1
- sglang/srt/models/mllama4.py +40 -4
- sglang/srt/models/opt.py +637 -0
- sglang/srt/models/qwen2_5_vl.py +29 -5
- sglang/srt/models/qwen2_audio.py +1 -1
- sglang/srt/models/qwen2_moe.py +120 -13
- sglang/srt/models/qwen2_vl.py +1 -1
- sglang/srt/models/qwen3.py +18 -3
- sglang/srt/models/qwen3_moe.py +32 -4
- sglang/srt/models/qwen3_next.py +1069 -0
- sglang/srt/models/qwen3_next_mtp.py +112 -0
- sglang/srt/models/qwen3_vl.py +787 -0
- sglang/srt/models/qwen3_vl_moe.py +471 -0
- sglang/srt/models/registry.py +15 -3
- sglang/srt/models/sarashina2_vision.py +269 -0
- sglang/srt/models/solar.py +505 -0
- sglang/srt/models/starcoder2.py +357 -0
- sglang/srt/models/step3_vl.py +1 -1
- sglang/srt/models/torch_native_llama.py +9 -2
- sglang/srt/models/utils.py +51 -0
- sglang/srt/multimodal/processors/base_processor.py +15 -7
- sglang/srt/multimodal/processors/dots_vlm.py +98 -0
- sglang/srt/multimodal/processors/glm4v.py +9 -9
- sglang/srt/multimodal/processors/internvl.py +153 -129
- sglang/srt/multimodal/processors/qwen_vl.py +23 -6
- sglang/srt/multimodal/processors/sarashina2_vision.py +81 -0
- sglang/srt/offloader.py +27 -3
- sglang/srt/parser/jinja_template_utils.py +6 -0
- sglang/srt/sampling/sampling_batch_info.py +38 -17
- sglang/srt/sampling/sampling_params.py +7 -0
- sglang/srt/server_args.py +966 -267
- sglang/srt/server_args_config_parser.py +146 -0
- sglang/srt/single_batch_overlap.py +151 -0
- sglang/srt/speculative/cpp_ngram/ngram.cpp +374 -0
- sglang/srt/speculative/cpp_ngram/ngram.h +110 -0
- sglang/srt/speculative/cpp_ngram/ngram_cache.py +138 -0
- sglang/srt/speculative/cpp_ngram/ngram_cache_binding.cpp +43 -0
- sglang/srt/speculative/cpp_ngram/param.h +125 -0
- sglang/srt/speculative/cpp_ngram/queue.h +71 -0
- sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +7 -1
- sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py +13 -2
- sglang/srt/speculative/{eagle_utils.py → eagle_info.py} +207 -757
- sglang/srt/speculative/eagle_worker.py +99 -28
- sglang/srt/speculative/ngram_utils.py +428 -0
- sglang/srt/speculative/ngram_worker.py +245 -0
- sglang/srt/speculative/spec_info.py +52 -0
- sglang/srt/speculative/spec_utils.py +606 -0
- sglang/srt/speculative/standalone_worker.py +109 -0
- sglang/srt/torch_memory_saver_adapter.py +5 -7
- sglang/srt/tracing/trace.py +578 -0
- sglang/srt/two_batch_overlap.py +8 -5
- sglang/srt/utils/__init__.py +2 -0
- sglang/srt/{utils.py → utils/common.py} +433 -77
- sglang/srt/{hf_transformers_utils.py → utils/hf_transformers_utils.py} +53 -5
- sglang/srt/{patch_torch.py → utils/patch_torch.py} +8 -0
- sglang/srt/utils/rpd_utils.py +452 -0
- sglang/srt/utils/slow_rank_detector.py +71 -0
- sglang/srt/warmup.py +8 -4
- sglang/srt/weight_sync/utils.py +2 -2
- sglang/test/attention/test_trtllm_mla_backend.py +169 -5
- sglang/test/get_logits_ut.py +57 -0
- sglang/test/run_eval.py +79 -11
- sglang/test/runners.py +5 -1
- sglang/test/simple_eval_common.py +5 -2
- sglang/test/simple_eval_mmmu_vlm.py +441 -0
- sglang/test/test_block_fp8.py +2 -2
- sglang/test/test_cutlass_moe.py +24 -6
- sglang/test/test_deterministic.py +297 -0
- sglang/test/test_disaggregation_utils.py +77 -0
- sglang/test/test_fp4_moe.py +370 -1
- sglang/test/test_programs.py +1 -1
- sglang/test/test_utils.py +383 -5
- sglang/utils.py +21 -1
- sglang/version.py +1 -1
- {sglang-0.5.2rc2.dist-info → sglang-0.5.3.dist-info}/METADATA +69 -124
- {sglang-0.5.2rc2.dist-info → sglang-0.5.3.dist-info}/RECORD +375 -245
- sglang/srt/disaggregation/launch_lb.py +0 -118
- sglang/srt/mem_cache/lora_radix_cache.py +0 -421
- /sglang/srt/{poll_based_barrier.py → utils/poll_based_barrier.py} +0 -0
- {sglang-0.5.2rc2.dist-info → sglang-0.5.3.dist-info}/WHEEL +0 -0
- {sglang-0.5.2rc2.dist-info → sglang-0.5.3.dist-info}/licenses/LICENSE +0 -0
- {sglang-0.5.2rc2.dist-info → sglang-0.5.3.dist-info}/top_level.txt +0 -0
@@ -1,15 +1,15 @@
|
|
1
1
|
sglang/__init__.py,sha256=CK60u1J439CJZaPihWqlDjFZ8doIC-5YgIYz-E1PIh4,1823
|
2
2
|
sglang/bench_offline_throughput.py,sha256=Bge0KG4Uiv9CHsxDnqKdXJmG_TcxM4blz3CmvLmDRvM,14815
|
3
|
-
sglang/bench_one_batch.py,sha256=
|
4
|
-
sglang/bench_one_batch_server.py,sha256=
|
5
|
-
sglang/bench_serving.py,sha256=
|
3
|
+
sglang/bench_one_batch.py,sha256=TFYI_wQGFpeJMV5r4C_EABmwZD3EDNuT1ORQX3jUWlY,23109
|
4
|
+
sglang/bench_one_batch_server.py,sha256=nD7GMOQE1I57BC1E7dTmqx-PMeSUPq8RF8W39s5z86I,26128
|
5
|
+
sglang/bench_serving.py,sha256=0l-QrhHtIMUVyewiuvxoRx6_MIulMpadkKtbzpQMxyQ,87841
|
6
6
|
sglang/check_env.py,sha256=qDMIG2rCNBH1yKnxQmF-Bp10oiFMUKMgfZLHZYOmdSY,8412
|
7
7
|
sglang/compile_deep_gemm.py,sha256=EsgboGA-MK3Rvx9TlUJPLxJ5LBira4bTcHJLKEp1H6k,6488
|
8
|
-
sglang/global_config.py,sha256=
|
9
|
-
sglang/launch_server.py,sha256=
|
10
|
-
sglang/profiler.py,sha256=
|
11
|
-
sglang/utils.py,sha256=
|
12
|
-
sglang/version.py,sha256=
|
8
|
+
sglang/global_config.py,sha256=qnMsK6AKzl8dhEbg-Jlbp_OVY9JUV0X9BqQiOl8VkiE,1737
|
9
|
+
sglang/launch_server.py,sha256=zBCKcUg4d7_-Zpp0NE9C7z8CIFYw94UH0wo7ZIilHMM,1063
|
10
|
+
sglang/profiler.py,sha256=zjD9-TxnDv3yU1Qmv4vItH5Nuhiu230ssqOA43EyH2g,4420
|
11
|
+
sglang/utils.py,sha256=YreOyt4-fWdDcg01abaMJ0y0ly4gYps26vRlmWhwGC8,17267
|
12
|
+
sglang/version.py,sha256=tgzuqHKcEdKBaP57F5oXxq4XlW2n9J4Fj8ZGu7nGOZg,22
|
13
13
|
sglang/eval/llama3_eval.py,sha256=mLNRZJIqV4CfqrY8UGnJEcHw2Xsyr1eyYZgFSUFYr1g,9997
|
14
14
|
sglang/eval/loogle_eval.py,sha256=-CC2s2kh5qUoDrHRkQVkC_jNvBgNojXbf456ny5s78s,4557
|
15
15
|
sglang/lang/api.py,sha256=rcp3GeoyZhmJ0GDLPRkuZNcxd0TBJy_wfUDpcmQoqW8,7210
|
@@ -23,87 +23,96 @@ sglang/lang/backend/anthropic.py,sha256=qXWqqFlk0-nC62QCcYIRjLTc4AHrdTKpl7hCeDzR
|
|
23
23
|
sglang/lang/backend/base_backend.py,sha256=tEutGCU5tni1xGlrIjAVvA9aflJBt2duKrIDzqabJWc,1975
|
24
24
|
sglang/lang/backend/litellm.py,sha256=ugmL7sfUxkUHVbHtwNzHgdQAEd4UCjNQboFuE3KThcY,2450
|
25
25
|
sglang/lang/backend/openai.py,sha256=YFs7pCDK6_DoYmjTa-vgDVdaeE4PYr7E-sIC1q7vAns,16422
|
26
|
-
sglang/lang/backend/runtime_endpoint.py,sha256=
|
26
|
+
sglang/lang/backend/runtime_endpoint.py,sha256=UZRKqQ8eE_N11oiv9rViF3THNq21A_uio0N0Bt9MaY8,17487
|
27
27
|
sglang/lang/backend/vertexai.py,sha256=gz0uNYyBb88jbPYz6ZIJ774fefrcbuVdoK33bphUZpI,4827
|
28
28
|
sglang/srt/_custom_ops.py,sha256=9GPNUyqnybgvUnzdQ8n5_NgluHmQNw0Gptos0iLPfrY,5342
|
29
29
|
sglang/srt/aio_rwlock.py,sha256=6LYtOdeTUY3hkfa1dmYkgsaF2ttrwIF3hUWz2AZ2fqw,2970
|
30
30
|
sglang/srt/bench_utils.py,sha256=zNTsTBUta27XxojgvR5p5X9M0v1MFzZ7dcX9TMq4NvQ,4348
|
31
31
|
sglang/srt/constants.py,sha256=0i-tEwG2BSYNDy96MxnGHV5HnBELkYcnsVGsE-R18o0,93
|
32
|
-
sglang/srt/custom_op.py,sha256=
|
33
|
-
sglang/srt/
|
32
|
+
sglang/srt/custom_op.py,sha256=m5F7x3LHlnOuesmcNmJtIWsfHGFLBJ807vKc05Gu1D0,3312
|
33
|
+
sglang/srt/environ.py,sha256=QaYZPUkvV-KyRWzJ6nz-kNO75Khn_pNbx2yX_YBL7-A,8894
|
34
34
|
sglang/srt/host_shared_memory.py,sha256=00TgVm3PNCl4rTPJZChG5uu4I5hd_Xf587r39tibmIg,2274
|
35
|
-
sglang/srt/offloader.py,sha256=
|
35
|
+
sglang/srt/offloader.py,sha256=cYON5jZErAp88Sh5JseaRw0B2Ho2MQz9smgvrTc-2KA,18765
|
36
36
|
sglang/srt/operations.py,sha256=za_h1_kcR1Te6-i_r0i_zAeumbmIRhLK2pS31sukkrw,6089
|
37
37
|
sglang/srt/operations_strategy.py,sha256=-w6GI2VgSGvORy3pUWRAJo6244ykSW1ZBV1Dx0fZlHA,7138
|
38
|
-
sglang/srt/
|
39
|
-
sglang/srt/
|
40
|
-
sglang/srt/
|
41
|
-
sglang/srt/torch_memory_saver_adapter.py,sha256=
|
42
|
-
sglang/srt/two_batch_overlap.py,sha256=
|
43
|
-
sglang/srt/
|
44
|
-
sglang/srt/
|
45
|
-
sglang/srt/
|
38
|
+
sglang/srt/server_args.py,sha256=pjtjMs62nnyMIRQMaL161QKDE2DViatyZNrfbYSkYuM,137969
|
39
|
+
sglang/srt/server_args_config_parser.py,sha256=niM5LYNaVxQg4FKHUJ3YmMZ-ekB3bDlvTwD7KqdLZoE,5220
|
40
|
+
sglang/srt/single_batch_overlap.py,sha256=m70hHl1ufIvc_9Nl2bYqnIcfoMIJkPyGjWIqJLBsKhI,4844
|
41
|
+
sglang/srt/torch_memory_saver_adapter.py,sha256=WFwPsOBvnWu6N2yTHmiOjLbDAYosslHwNSfyFHjz-7w,2511
|
42
|
+
sglang/srt/two_batch_overlap.py,sha256=0cHUV2ZeyCaojDRpqNcRSUIQM3_Y4H4HfssML2gt_6k,34240
|
43
|
+
sglang/srt/warmup.py,sha256=fQYzNp-I9mZccs2BerI6mOZd1CDQjdiuFCeUkcwiOb4,1834
|
44
|
+
sglang/srt/batch_invariant_ops/__init__.py,sha256=Z9TvOE4BL7cAaBIePEsVglH7RcWBSBIsSHs-XbveQNE,728
|
45
|
+
sglang/srt/batch_invariant_ops/batch_invariant_ops.py,sha256=5vtvEFezea4tjGFuVMkadgZxynxAZZ8yxGwpkSm7Sl4,16246
|
46
|
+
sglang/srt/configs/__init__.py,sha256=aMxLOh72NAL0iQl0aPcPP2Ih9oY5vhHUHAcymE0fv-4,1142
|
46
47
|
sglang/srt/configs/chatglm.py,sha256=j-b0YkdYUmQm2y1kNmMJtKeACxWKmBbvNNkDWbs6kbI,2907
|
47
48
|
sglang/srt/configs/dbrx.py,sha256=tdhIkXAQl1yr0MxqFmsDG1E0e2puRTTKm6UTyANBLac,11005
|
48
49
|
sglang/srt/configs/deepseekvl2.py,sha256=sYMmoy5_LlDXht7MTi4QGYOaiE8iMGUO44bnuOu4i8I,23683
|
49
|
-
sglang/srt/configs/device_config.py,sha256=
|
50
|
+
sglang/srt/configs/device_config.py,sha256=1Iaxrk65bJsCWV3E2cDfiap_T_xJIIj9oFVn9y6v0EI,515
|
51
|
+
sglang/srt/configs/dots_ocr.py,sha256=HSKerqzHzrKIzoNZOUXA_KiBLmL-Rs8h4Vw17pjF1Uo,1863
|
52
|
+
sglang/srt/configs/dots_vlm.py,sha256=PYoSrT-sRoykfNhdXFauqv21Sb54iavjYD3F09xww3U,5176
|
50
53
|
sglang/srt/configs/exaone.py,sha256=Duxd4yQoKy8GWEzZD_kCY_OzmN_67CTJL_Kgn0eXk3g,10731
|
54
|
+
sglang/srt/configs/falcon_h1.py,sha256=qn_n2uLRVN4EK4gn4UoHo_1NhV6kPFdOIb5TBJ0NAYs,16395
|
51
55
|
sglang/srt/configs/internvl.py,sha256=rIC-CBWJHy-Mjmk0_H3YcvVmZzeCRbXGo3dQMex-yeg,28141
|
52
56
|
sglang/srt/configs/janus_pro.py,sha256=yI2c_aWBIKiTcK6RZxK6hq8CIxyYRH_NN8uaSNKsrKI,19156
|
53
57
|
sglang/srt/configs/kimi_vl.py,sha256=4W7VQI3pr888ZsFA2SqCQo4mI0seXTOrGQ-x3oTvWew,1358
|
54
58
|
sglang/srt/configs/kimi_vl_moonvit.py,sha256=hx2Rt4JSFbvy2HUTeLjBpge87m8M6ITAhqsgdNf_Jd4,1163
|
55
|
-
sglang/srt/configs/load_config.py,sha256=
|
59
|
+
sglang/srt/configs/load_config.py,sha256=_XhZdoP2Kbzih3FBZiiwGhXBf2-bNrn6f8eA63zqS7Y,3840
|
56
60
|
sglang/srt/configs/longcat_flash.py,sha256=Qp25xJVLq2K72Z80cXhcJxtqhagAdiPySDoevuT0Sno,3589
|
57
|
-
sglang/srt/configs/model_config.py,sha256=
|
61
|
+
sglang/srt/configs/model_config.py,sha256=fICFIXWjFrS7xtQYA-8TNJigD9blBfnEKmsbDsbMenE,35245
|
62
|
+
sglang/srt/configs/qwen3_next.py,sha256=MVMDzdwlxrdar_6p6n9HjaXAei_giFfXlQNJAWpSLQ4,15794
|
63
|
+
sglang/srt/configs/qwen3_vl.py,sha256=FMC-SVeM_sxLdjEAfJ_oNog1P5kXm70NMHNXmzSMo_U,29008
|
58
64
|
sglang/srt/configs/step3_vl.py,sha256=_Otgnym57DVgB_kZ__8c1_Ys5gSalA_K0ZuVjcG51T0,4845
|
59
65
|
sglang/srt/configs/update_config.py,sha256=GEf-XhL8JPrbX9-Hz8V7S3M6YTg76DVdIhc_4YdMDtc,6291
|
60
66
|
sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
|
61
|
-
sglang/srt/connector/__init__.py,sha256=
|
67
|
+
sglang/srt/connector/__init__.py,sha256=zPuaRKulJsrDyT3OtN9wUT-4fqQBX-cJr2fZUB0_KPM,1562
|
62
68
|
sglang/srt/connector/base_connector.py,sha256=LCKsTX_yWHclPrzNNB1fLwVCN8rIDKZaZhqHQ54B9MU,2774
|
63
69
|
sglang/srt/connector/redis.py,sha256=K3fJSm2rZsd0R_KkICV3jIWKtrwxXE1tYktWnovLbz4,2514
|
70
|
+
sglang/srt/connector/remote_instance.py,sha256=X_PIZFvZPsnUZ_54AgnzR0NdUfi6ClwQVlNexZatkXU,2710
|
64
71
|
sglang/srt/connector/s3.py,sha256=hn5Ow9Y6ItS8RMJDwnTQe5QaiQ7HKMgiwggE9QnUewM,3743
|
65
72
|
sglang/srt/connector/utils.py,sha256=isTvxauz1_8no5MW6p8Bwj2H9mQkweaRO_RSwAPA7R8,983
|
66
73
|
sglang/srt/connector/serde/__init__.py,sha256=zuleyYN-jrDyiPiXjTbQKA27llb4H7HuDTZcBwrLk1U,707
|
67
74
|
sglang/srt/connector/serde/safe_serde.py,sha256=XSjb9mRv1HqeG6Jiw633EOQYlrcI23Q21UluYqPZaWA,760
|
68
75
|
sglang/srt/connector/serde/serde.py,sha256=n59I2MXLa7WCyN_8pEd8L-scJk7lMhmEX-GOUIhF0ZA,1004
|
69
|
-
sglang/srt/constrained/base_grammar_backend.py,sha256=
|
70
|
-
sglang/srt/constrained/llguidance_backend.py,sha256=
|
71
|
-
sglang/srt/constrained/outlines_backend.py,sha256=
|
72
|
-
sglang/srt/constrained/outlines_jump_forward.py,sha256=
|
76
|
+
sglang/srt/constrained/base_grammar_backend.py,sha256=FKlBqllZ-bs65DYG5l-qsuuqfXOPeuUe5_NvPlpBzUs,8114
|
77
|
+
sglang/srt/constrained/llguidance_backend.py,sha256=JkA_59nZ4zd8ciGvXsZ-OEsGCyQMoIxBO8QXIUqAUn4,6161
|
78
|
+
sglang/srt/constrained/outlines_backend.py,sha256=t4gkO9AX8xfYulvhNx0n-p1kUavJnCjE9XjwN1Z4BV4,6883
|
79
|
+
sglang/srt/constrained/outlines_jump_forward.py,sha256=dm3PVswx2oE_JfAv-tPd7irMI95VCq4cZT-AF-Ohx3s,6823
|
73
80
|
sglang/srt/constrained/reasoner_grammar_backend.py,sha256=YFxWuOTTo4e6cGhxnaBwuwli6f8FTUJtszib8dBq_8I,3207
|
74
|
-
sglang/srt/constrained/xgrammar_backend.py,sha256=
|
81
|
+
sglang/srt/constrained/xgrammar_backend.py,sha256=Dhn0sROXQtybY-sJfMcleIu_Rmn0sjleGMywCD3Simk,9338
|
75
82
|
sglang/srt/constrained/triton_ops/bitmask_ops.py,sha256=WjTen9iuuFWLzkE1mAHQZB9_7aIy5QH8Wjf-lB-Fams,4614
|
76
83
|
sglang/srt/debug_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
77
|
-
sglang/srt/debug_utils/dump_comparator.py,sha256=
|
78
|
-
sglang/srt/debug_utils/
|
79
|
-
sglang/srt/debug_utils/
|
80
|
-
sglang/srt/
|
81
|
-
sglang/srt/disaggregation/
|
84
|
+
sglang/srt/debug_utils/dump_comparator.py,sha256=3rWdv4BQXTDtB3z5ao9VPN5lK31ImKf2zMWX7f0Xe6s,5394
|
85
|
+
sglang/srt/debug_utils/dump_loader.py,sha256=RzWpl7RsDRw98DJMEvf6U9c9fPiobyeWI8opmRaRGGA,2632
|
86
|
+
sglang/srt/debug_utils/dumper.py,sha256=6frfI4WErneQE4RpXgAR5gZJQRdzcQjyO3McW56jgYo,3517
|
87
|
+
sglang/srt/debug_utils/text_comparator.py,sha256=NLJdjUikW--bWnGQWIeFJesyuvZ2Vni2S53fIED-m8s,7370
|
88
|
+
sglang/srt/disaggregation/decode.py,sha256=_k9uVKzFUxEJIpBQSjoBVaY-GMoc0zTW6FuVhiH05os,36936
|
89
|
+
sglang/srt/disaggregation/decode_kvcache_offload_manager.py,sha256=BWjhO0CDdpU8VIC_-hL9R8CjqUk5cQ1ivfXLjPSS9jU,7029
|
90
|
+
sglang/srt/disaggregation/decode_schedule_batch_mixin.py,sha256=BeGxXbFLW_ZbuyOs0idWzlicCh6NA1pTOXgON5SAdaE,6947
|
82
91
|
sglang/srt/disaggregation/kv_events.py,sha256=yFbtwOeblcCmOXTjg00TopxnyjkmCBQIVz46KB8jetY,13555
|
83
|
-
sglang/srt/disaggregation/
|
84
|
-
sglang/srt/disaggregation/
|
85
|
-
sglang/srt/disaggregation/
|
86
|
-
sglang/srt/disaggregation/utils.py,sha256=evdv4AkQxJzTqdO0jrdAr7vHUbBSCEXMzNREh0FoRLc,12294
|
92
|
+
sglang/srt/disaggregation/mini_lb.py,sha256=JY2uCeUM3Zk3FdGWmCQBY3LeiJ5FM1PlrqAW8wJTTHI,395
|
93
|
+
sglang/srt/disaggregation/prefill.py,sha256=eOEdepYknYHO6vz-MCV-jmuTOe_qq9X_ZD_AX1SwICU,37138
|
94
|
+
sglang/srt/disaggregation/utils.py,sha256=oBMqYOMAnyK4T69fFaDIuM1AAGa2vOvoA-CBtGocOkk,12263
|
87
95
|
sglang/srt/disaggregation/ascend/__init__.py,sha256=-lxnica6fZYmYNpKwmE8yESUpQ5Sxf2DiZoBHzboKc8,146
|
88
|
-
sglang/srt/disaggregation/ascend/conn.py,sha256=
|
89
|
-
sglang/srt/disaggregation/ascend/transfer_engine.py,sha256=
|
96
|
+
sglang/srt/disaggregation/ascend/conn.py,sha256=Z0xTJjSYXal7uGA5e7ci4oxOcWdC9qrV-gMd06ikp2w,4109
|
97
|
+
sglang/srt/disaggregation/ascend/transfer_engine.py,sha256=JkalkQUFCwdCjJ_k8hxJqmgJ4CfxgiQzHzplP1S6xqE,3577
|
90
98
|
sglang/srt/disaggregation/base/__init__.py,sha256=4VwUv0aWxwmVL1049XK82aLTNxmt0WY5RPy9li-wyVk,160
|
91
|
-
sglang/srt/disaggregation/base/conn.py,sha256=
|
99
|
+
sglang/srt/disaggregation/base/conn.py,sha256=0CGEv-B24woyDBdEgQwT8Gqz10Uox3gCgWiCxznXkwQ,2912
|
92
100
|
sglang/srt/disaggregation/common/__init__.py,sha256=7yl-EGLMVKRpBUaGF_7lwAsw2J_mqpRZV0238VGxD9o,126
|
93
|
-
sglang/srt/disaggregation/common/conn.py,sha256=
|
101
|
+
sglang/srt/disaggregation/common/conn.py,sha256=PiYuKGYVeVxoPaQcUK673IGelN9XnK7fyHSb6y6mptY,24415
|
94
102
|
sglang/srt/disaggregation/common/utils.py,sha256=SxRhAWisNK8seGhb5BXBJ5u53DF7yeKVPMWPcB5ywbE,1194
|
95
103
|
sglang/srt/disaggregation/fake/__init__.py,sha256=jJGWdXwaQiGIoR6atKqkQfkJmVyQ09l55VUN2WjwaeY,77
|
96
|
-
sglang/srt/disaggregation/fake/conn.py,sha256=
|
104
|
+
sglang/srt/disaggregation/fake/conn.py,sha256=qsN5SEQLMenALxYC3x7GYZprhUQyRje5ldNwjkT6OPw,2283
|
97
105
|
sglang/srt/disaggregation/mooncake/__init__.py,sha256=0TgqkAdQI1YynbHY6c0QISvVoOSk-0SwCIq5rjPSmgE,156
|
98
|
-
sglang/srt/disaggregation/mooncake/conn.py,sha256=
|
106
|
+
sglang/srt/disaggregation/mooncake/conn.py,sha256=01TcfkekA3v7ApN11ATx_YsFi6B5U1xd3yPDxRx6F9Y,50697
|
99
107
|
sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=O-XRmQ3QwD6RuiyRgw4ZohSrFkuPdppV_eDBMUXn0jk,5884
|
100
108
|
sglang/srt/disaggregation/nixl/__init__.py,sha256=qODVPIGWUXKXq4zsRIcMYoAoAeg6nBIN9vdQOlVMANE,136
|
101
|
-
sglang/srt/disaggregation/nixl/conn.py,sha256=
|
109
|
+
sglang/srt/disaggregation/nixl/conn.py,sha256=_si9ryeRoDHYmomthSyztUOEYl20aa18mS9_Sm961S4,32299
|
102
110
|
sglang/srt/distributed/__init__.py,sha256=jFOcyt-wFAPMBUAf9zkZalNQlt-4rqmT6pCKBz1E4qo,149
|
103
111
|
sglang/srt/distributed/communication_op.py,sha256=IBnFUdMftK_VSTMMMitGveonorFUUVNL4guqO31cMSc,1130
|
104
112
|
sglang/srt/distributed/naive_distributed.py,sha256=5Kcfapzz61G3TtScTZrHoWa4bf6Vr27GlMcBAGMz7tQ,3260
|
105
|
-
sglang/srt/distributed/parallel_state.py,sha256=
|
113
|
+
sglang/srt/distributed/parallel_state.py,sha256=m0T-Tfqi7DnCUPEGhA8Fr-dNlbeoowvG3Ds60IoDeos,68541
|
106
114
|
sglang/srt/distributed/utils.py,sha256=aaCxATncLGnVgB0WlGpBdee0behKW8Dy_dakqcuKSaQ,8497
|
115
|
+
sglang/srt/distributed/device_communicators/all_reduce_utils.py,sha256=IxCuujkRVUGckr-PAhe2F_QfHGKhvQHo9ryB9-CQjpA,332
|
107
116
|
sglang/srt/distributed/device_communicators/cuda_wrapper.py,sha256=3jvPG-Ow5UBLiXhfx8T8snR7crSZbPpARAggsDPWq7k,7038
|
108
117
|
sglang/srt/distributed/device_communicators/custom_all_reduce.py,sha256=Q1kkKPKFPV0QMmKLyjOBlOnX8-Pr4UeGBZYkG6j0gc0,16570
|
109
118
|
sglang/srt/distributed/device_communicators/custom_all_reduce_utils.py,sha256=Fx9D70NO2dplyuqOxTasMwkzONN3wfC4WOny3tWSbPA,14159
|
@@ -114,122 +123,170 @@ sglang/srt/distributed/device_communicators/pynccl.py,sha256=7aE-pKMyJfawGo7Vzym
|
|
114
123
|
sglang/srt/distributed/device_communicators/pynccl_allocator.py,sha256=zAxWwI73ZNsF_c4PZUnRMzwzSD88v26w8komzZ4Uatc,4771
|
115
124
|
sglang/srt/distributed/device_communicators/pynccl_wrapper.py,sha256=BLGEl90GZSl3K-04x7eCTaVmpZBXpcb7uJakAseteFU,18539
|
116
125
|
sglang/srt/distributed/device_communicators/quick_all_reduce.py,sha256=4j1_E4azoxfd8wxtfFmt9rvbQncl8ny6wmTMl6gAkp0,9932
|
117
|
-
sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=
|
126
|
+
sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=q68niUgh8zF7CgNwqny9vaZU1aHGpLSrxBMiUWZ5Nsk,20969
|
127
|
+
sglang/srt/distributed/device_communicators/symm_mem.py,sha256=EFN6v2l3Hrxek_1kxDHMMmzLP-bDswhdNW53FrQm2-I,5664
|
118
128
|
sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
|
119
129
|
sglang/srt/entrypoints/EngineBase.py,sha256=yKN76witT2jz1zhmLHmPNLGMpK2UiOTaKQ2KPD8l99U,2594
|
120
130
|
sglang/srt/entrypoints/context.py,sha256=aD-94xkD0komuGO5gtYUoJKCHdc4hAipMxQt04yVRGA,8030
|
121
|
-
sglang/srt/entrypoints/engine.py,sha256=
|
131
|
+
sglang/srt/entrypoints/engine.py,sha256=hDuQJ5XheUZNfScFRBxKHEev_gDTiKH3f6gKtcWfj-U,34950
|
132
|
+
sglang/srt/entrypoints/grpc_request_manager.py,sha256=0tVlP5eRgQTk1PGWsQtzNrlqd3-HJp67yu3o6jRkan4,32273
|
133
|
+
sglang/srt/entrypoints/grpc_server.py,sha256=WjToYISgaWQwlmUz6-1PxMd84YPGmTPpbdKvNiukt4A,31208
|
122
134
|
sglang/srt/entrypoints/harmony_utils.py,sha256=01T-A5GBUm2b306PcxNEg2rfx4cykBcqNYrzcXTWBlc,13590
|
123
|
-
sglang/srt/entrypoints/http_server.py,sha256=
|
135
|
+
sglang/srt/entrypoints/http_server.py,sha256=8f9fJhaXGEGwMCsLp5B2Ukk_MDd5dNRYE8NJ5FbLKk0,52423
|
124
136
|
sglang/srt/entrypoints/http_server_engine.py,sha256=_--j4U04OeJLlnnv1f0XmCd_Ry0z1FlhkrbePX8rYV0,4938
|
125
137
|
sglang/srt/entrypoints/tool.py,sha256=aAXEjBqkr5j6_IgRL8fZCKgoDPY2TikM20-LQXU9TaI,2709
|
126
138
|
sglang/srt/entrypoints/openai/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
127
|
-
sglang/srt/entrypoints/openai/protocol.py,sha256=
|
128
|
-
sglang/srt/entrypoints/openai/serving_base.py,sha256=
|
129
|
-
sglang/srt/entrypoints/openai/serving_chat.py,sha256=
|
130
|
-
sglang/srt/entrypoints/openai/serving_completions.py,sha256=
|
131
|
-
sglang/srt/entrypoints/openai/serving_embedding.py,sha256=
|
132
|
-
sglang/srt/entrypoints/openai/serving_rerank.py,sha256=
|
133
|
-
sglang/srt/entrypoints/openai/serving_responses.py,sha256=
|
134
|
-
sglang/srt/entrypoints/openai/serving_score.py,sha256=
|
139
|
+
sglang/srt/entrypoints/openai/protocol.py,sha256=UmPdjkfaFYP2X5NYI0rt8L1w2pdqGYsIcIiCO1bnF7Y,33216
|
140
|
+
sglang/srt/entrypoints/openai/serving_base.py,sha256=b15N627nn3me2jHzXJBFkWPpX8OJdtD66v8aSsx0i0s,7746
|
141
|
+
sglang/srt/entrypoints/openai/serving_chat.py,sha256=w1awHs-msdhKEzlMHUA_e_Ptx3qlUcdiUGVLq6ryYfs,49829
|
142
|
+
sglang/srt/entrypoints/openai/serving_completions.py,sha256=SBoaa9KIIL7NQCWwB1B2A9roTrQW1P9C4fkhuHymo00,18233
|
143
|
+
sglang/srt/entrypoints/openai/serving_embedding.py,sha256=H7Nv1vSUuyOUXBeaNvYB4XMFxmcJc8or7MUKyuAw_XA,6435
|
144
|
+
sglang/srt/entrypoints/openai/serving_rerank.py,sha256=G79XdomjzLhH1YCbGR4alSxWO9NNmi7U-YSPsH61cjk,3379
|
145
|
+
sglang/srt/entrypoints/openai/serving_responses.py,sha256=ro1VwdazJNqm5RRVbMKI7NliPsVMFcrNEsRD23QiADc,54989
|
146
|
+
sglang/srt/entrypoints/openai/serving_score.py,sha256=LuwXzWRAMm-xMK8ZjP3JaJ84Gxjr0Oi-e3keCsIgK38,1992
|
135
147
|
sglang/srt/entrypoints/openai/tool_server.py,sha256=rRNGELPV5RHONp55MYkO4_b0NPsPffJnWg-gMYwDSTc,5659
|
136
148
|
sglang/srt/entrypoints/openai/usage_processor.py,sha256=9LTB5rqdRuMKyZrIXiUBuF_WKaSg9X45YdzERDxbtCY,2746
|
137
149
|
sglang/srt/entrypoints/openai/utils.py,sha256=DT9quW0F1DX1P9NAuMxpERgXcNYKKtm7rd7J1m7EaDQ,2112
|
138
150
|
sglang/srt/eplb/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
139
|
-
sglang/srt/eplb/eplb_manager.py,sha256=
|
140
|
-
sglang/srt/eplb/expert_distribution.py,sha256=
|
141
|
-
sglang/srt/eplb/expert_location.py,sha256=
|
151
|
+
sglang/srt/eplb/eplb_manager.py,sha256=e8HMGzZb2QFRGdS7JZIFSy8cOIlxOlGRY-ef7Wei0Sw,4364
|
152
|
+
sglang/srt/eplb/expert_distribution.py,sha256=gd3PktGCSqHLjB5XHhTSA-OtaDVYw-taExCe7NOHM9E,32939
|
153
|
+
sglang/srt/eplb/expert_location.py,sha256=WoOmsZ4u9le44km3nbjJjF6BVBxN3cKRz9nQM0uWH_s,18386
|
142
154
|
sglang/srt/eplb/expert_location_dispatch.py,sha256=vA0o-7bXf847BQAVCqElWKEpofZLAIsir_d0pew2KNQ,4178
|
143
|
-
sglang/srt/eplb/expert_location_updater.py,sha256=
|
155
|
+
sglang/srt/eplb/expert_location_updater.py,sha256=ScoyCJbtJNQLriQXJNZTV2-5G8-tR3AocZDK3ujMaJk,21172
|
144
156
|
sglang/srt/eplb/eplb_algorithms/__init__.py,sha256=c6fo4qM8N0p5DyyfGp9fxqc2wkxvzSJBzJFRG1I1leg,1959
|
145
157
|
sglang/srt/eplb/eplb_algorithms/deepseek.py,sha256=mMZT7zAAArccdRS0xXxifvMb3qn9enSt426uUTKeiq4,8340
|
146
158
|
sglang/srt/eplb/eplb_algorithms/deepseek_vec.py,sha256=Vzy5Iarua1VgVHzjBNZaVV_vt1LY1BLtJz7PmzB701k,10654
|
147
159
|
sglang/srt/eplb/eplb_simulator/__init__.py,sha256=HIZaXqsvsOCMMJ81dSs1afSGZ5G6OiOZEtHpltyzzjY,21
|
148
160
|
sglang/srt/eplb/eplb_simulator/reader.py,sha256=OlsWqsuXqVQY7PfvvrP-vsl2Ww2Kg4vD_K9BltM_QHo,1828
|
149
|
-
sglang/srt/function_call/base_format_detector.py,sha256=
|
161
|
+
sglang/srt/function_call/base_format_detector.py,sha256=DunwyYJxLcwrD-iHDXX9uqJ5S8zMJhUXfzHiBgbXaZ0,15728
|
150
162
|
sglang/srt/function_call/core_types.py,sha256=wLUsW8mUA-i-ISz5QUDL7Ejq72W-K1HIhFa9Wm-_oA8,786
|
151
163
|
sglang/srt/function_call/deepseekv31_detector.py,sha256=hpRjjv219mH0v1PQpMvj_gezrP7PwvLiNbBKOTCSC3U,9731
|
152
164
|
sglang/srt/function_call/deepseekv3_detector.py,sha256=DhO3tUX1nvFlfRkHJ1Vm_7OMXFFTvWBA9u9SZKFAwuQ,9710
|
153
|
-
sglang/srt/function_call/ebnf_composer.py,sha256=
|
154
|
-
sglang/srt/function_call/function_call_parser.py,sha256=
|
155
|
-
sglang/srt/function_call/glm4_moe_detector.py,sha256=
|
156
|
-
sglang/srt/function_call/gpt_oss_detector.py,sha256=
|
157
|
-
sglang/srt/function_call/
|
165
|
+
sglang/srt/function_call/ebnf_composer.py,sha256=OpjfvYowLlMLnxz5fwS7zv5-d43_ZwO9ucGAMsILvm4,15321
|
166
|
+
sglang/srt/function_call/function_call_parser.py,sha256=NUQCh7dMwRTkudRDFz3Mdr-nDL_zXqABVQYPocGI274,8592
|
167
|
+
sglang/srt/function_call/glm4_moe_detector.py,sha256=fUqozcc_V2g4Vd0l3BjhZZ0MRKmLinx98-Gm8aOlVgI,6975
|
168
|
+
sglang/srt/function_call/gpt_oss_detector.py,sha256=f0fjnxNKLb_TrxvNWbmCOB3_KN07NtZacazFpyIcqC8,9262
|
169
|
+
sglang/srt/function_call/json_array_parser.py,sha256=i985g7MUHj1M8gJ0it3Lw663Bx1uh0biN_It13hnMxU,2227
|
170
|
+
sglang/srt/function_call/kimik2_detector.py,sha256=VWN4Nf9l9O0uSUR0UppuoAyVObvGw-rVvklFJ1qzbRA,10641
|
158
171
|
sglang/srt/function_call/llama32_detector.py,sha256=trZja7IXc2IWUko5PRj7p4pRU6XCVjtxr-7qJNr0EWo,3647
|
159
172
|
sglang/srt/function_call/mistral_detector.py,sha256=Ts2HrJdu5r2lCCMVKPZfnCds9pt1K5H1EaOQrF_HD60,4686
|
160
173
|
sglang/srt/function_call/pythonic_detector.py,sha256=yAatItKtVDf9uhlB78wPWKP3fbefLJDNGGzFV5ey-YA,8756
|
161
174
|
sglang/srt/function_call/qwen25_detector.py,sha256=nv83_sipZMU6Vie3J4HeNiZVCSRU5ei7601j50779HA,5181
|
162
|
-
sglang/srt/function_call/qwen3_coder_detector.py,sha256=
|
175
|
+
sglang/srt/function_call/qwen3_coder_detector.py,sha256=fyiF0WkNwozlKZQy0G903hj-ZAyvEgkFC8l20FVOVX8,14299
|
163
176
|
sglang/srt/function_call/step3_detector.py,sha256=SAibrispu0peDQzqOG2-V93KGHtV6Ud7a38vlnhTunA,17442
|
164
|
-
sglang/srt/function_call/utils.py,sha256=
|
165
|
-
sglang/srt/
|
177
|
+
sglang/srt/function_call/utils.py,sha256=HeRwtWRrkwipIKM_EBfCgyNxy2It_7kAfPUcBhMERjU,4548
|
178
|
+
sglang/srt/grpc/__init__.py,sha256=dPeDsMRMW8WyHD6jolIpIlvlJsYLAhwewHhCTdu1MAE,21
|
179
|
+
sglang/srt/grpc/compile_proto.py,sha256=kiaVgK1eEcSMVd5mCed-5cs_78JIawjJ1jQ2U6mdYkg,6997
|
180
|
+
sglang/srt/grpc/sglang_scheduler_pb2.py,sha256=Q0QVnMvRovwWpFmd3j2-fNLZfuvX_4Hq8iRKM6sz-s4,14182
|
181
|
+
sglang/srt/grpc/sglang_scheduler_pb2.pyi,sha256=RN5UOWYzxPBFUUcWAaiXPeOB9iclizo7ehFzNrI5P94,23095
|
182
|
+
sglang/srt/grpc/sglang_scheduler_pb2_grpc.py,sha256=s8WXOFE0a0zGkcfVUxHxa28yA8VOZxua5K31dEXqt_A,9051
|
183
|
+
sglang/srt/layers/activation.py,sha256=iWEi9wIdAf0xov0KU27JUB5FIl6dqCGQjbIm_9dVSfQ,13368
|
166
184
|
sglang/srt/layers/amx_utils.py,sha256=1mENgHK2B8mgaD1oMtgbZ15Jmy_Uu1QueBmo09Ff2iA,2865
|
167
|
-
sglang/srt/layers/communicator.py,sha256=
|
168
|
-
sglang/srt/layers/dp_attention.py,sha256=
|
169
|
-
sglang/srt/layers/elementwise.py,sha256=
|
185
|
+
sglang/srt/layers/communicator.py,sha256=KzmCXRknuxlRnwdA3G38wRlVHYD_kf76mOZWhaIU5lo,24366
|
186
|
+
sglang/srt/layers/dp_attention.py,sha256=ly-EMgb_I0neGA0Y4Hq9Z_sp1plIRwoBM3feTqLJob4,16322
|
187
|
+
sglang/srt/layers/elementwise.py,sha256=tGbN7bbK2s-j3DoJeJfkfyPNmKdVAutAd0J3TbxUPAQ,18849
|
170
188
|
sglang/srt/layers/flashinfer_comm_fusion.py,sha256=hnVVS_VsWukTYhREubWs2KcJ1GjVMEI9ECqxqOMXZv8,6729
|
171
|
-
sglang/srt/layers/layernorm.py,sha256=
|
172
|
-
sglang/srt/layers/linear.py,sha256=
|
173
|
-
sglang/srt/layers/logits_processor.py,sha256=
|
189
|
+
sglang/srt/layers/layernorm.py,sha256=WKAv4N6LFDLu9x62XBqGT9xEcrj2W53GtP8ji4zTbTM,11335
|
190
|
+
sglang/srt/layers/linear.py,sha256=z15L32URxbPUb34_urGDwSWIZNfh1jmyKUF1A9lzI9I,55761
|
191
|
+
sglang/srt/layers/logits_processor.py,sha256=BYDSuqVCrpo67JFCSvmATfegVwgtJ-2_Bgwp23y-Ds8,28094
|
174
192
|
sglang/srt/layers/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
|
175
193
|
sglang/srt/layers/multimodal.py,sha256=vNswe0J0maeBiN8ehlmxfs-YYlrmMllYl6Nc_Y57Neg,5114
|
176
|
-
sglang/srt/layers/parameter.py,sha256=
|
194
|
+
sglang/srt/layers/parameter.py,sha256=6ce1lx2urWJKz87g2qg_oce_S5BV0IcguSiOGGHbcC8,18378
|
177
195
|
sglang/srt/layers/pooler.py,sha256=uZ6WX1FLMEafZwusyZdm6KuVlIwSjbKrdwk2qzgqNGk,3812
|
178
196
|
sglang/srt/layers/radix_attention.py,sha256=MpugwaqiY2zPTeg7FVW41OljljCXmYVvHZ2MvneF6PI,3745
|
179
197
|
sglang/srt/layers/rocm_linear_utils.py,sha256=7PPh-5H2IYjiLp_g_szzVBsmiMacZ1RtjVtjdrdFJKI,1374
|
180
|
-
sglang/srt/layers/rotary_embedding.py,sha256=
|
181
|
-
sglang/srt/layers/sampler.py,sha256=
|
198
|
+
sglang/srt/layers/rotary_embedding.py,sha256=D6X_HmwtGShR_XKuOBPXtRSmfrUU_VMuJ58HycmRUxY,75058
|
199
|
+
sglang/srt/layers/sampler.py,sha256=C4iNOK42sfUGKhA9xBZA4xXHDsciC3V_5aPoJr-kpJM,19805
|
182
200
|
sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSahyMfA,4063
|
183
|
-
sglang/srt/layers/utils.py,sha256=
|
201
|
+
sglang/srt/layers/utils.py,sha256=58yO8aKsAvS0jJ3nhOWVR_WAc5oTFiMeAERWNkXGR-I,1945
|
184
202
|
sglang/srt/layers/vocab_parallel_embedding.py,sha256=x0421kNUgnAb5LnkJR6O67vGZ0ikSD-ThluV3QGsZ_8,22673
|
185
|
-
sglang/srt/layers/attention/aiter_backend.py,sha256=
|
186
|
-
sglang/srt/layers/attention/ascend_backend.py,sha256=
|
187
|
-
sglang/srt/layers/attention/
|
188
|
-
sglang/srt/layers/attention/
|
203
|
+
sglang/srt/layers/attention/aiter_backend.py,sha256=oovonzjvnijaegNmkyTlWlEhCuhXAMNd85EykvKvvow,43923
|
204
|
+
sglang/srt/layers/attention/ascend_backend.py,sha256=JwbuW4a-urczuVHEP7oRGOehJOcfUO71LCG3BiQq-zA,27206
|
205
|
+
sglang/srt/layers/attention/attention_registry.py,sha256=RT9Z1Jg0pM65FO08S9Vs3K_7BxfOYpFsgbl3_FZ-uZ4,6795
|
206
|
+
sglang/srt/layers/attention/base_attn_backend.py,sha256=AGCPd_VqvvZ0tLExCEs_9RenwhP7nB47nNcbR9VuSJw,3831
|
207
|
+
sglang/srt/layers/attention/cutlass_mla_backend.py,sha256=cb2dMTnDMmMVBG_lqFRciVDiF3PeEAcXYaKDFWszeTw,9849
|
189
208
|
sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
|
190
|
-
sglang/srt/layers/attention/dual_chunk_flashattention_backend.py,sha256=
|
191
|
-
sglang/srt/layers/attention/flashattention_backend.py,sha256=
|
192
|
-
sglang/srt/layers/attention/flashinfer_backend.py,sha256=
|
193
|
-
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=
|
194
|
-
sglang/srt/layers/attention/flashmla_backend.py,sha256=
|
195
|
-
sglang/srt/layers/attention/hybrid_attn_backend.py,sha256=
|
196
|
-
sglang/srt/layers/attention/
|
209
|
+
sglang/srt/layers/attention/dual_chunk_flashattention_backend.py,sha256=20vmnq68-Cq62UWIzQU_uwJzq4B8KdJeZUPVoWPiyDs,68620
|
210
|
+
sglang/srt/layers/attention/flashattention_backend.py,sha256=nuOTQ8Zl2anCrNYDLoDiBhCkd3XE4UAA9ZThcY7zFOU,105158
|
211
|
+
sglang/srt/layers/attention/flashinfer_backend.py,sha256=3tX5ThHhH4Za0jhONXxxViUxN1sYmuqkdOOD0TGq9Go,52247
|
212
|
+
sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=hEoTNUHPM1LJXMPNVY9FGazUnord_Bmr5WsQY0mxRx0,39403
|
213
|
+
sglang/srt/layers/attention/flashmla_backend.py,sha256=QqQgWIcF90wVqcvsxJR7lHtCtAqkNd9Rdl-tLq7cHHk,20852
|
214
|
+
sglang/srt/layers/attention/hybrid_attn_backend.py,sha256=Mp1M83ODwHENVvefP2JEjv47ifDt5SA1hoyrJo4OJyU,5128
|
215
|
+
sglang/srt/layers/attention/hybrid_linear_attn_backend.py,sha256=e8O2h82ZDwgDEXt8QF23IznOY3vq1-LB9vcWum7Y68k,21938
|
216
|
+
sglang/srt/layers/attention/intel_amx_backend.py,sha256=IU_E7HkPrBLfNG2ArhI1ozHZ_pROujUzdI6g2JuToL0,4154
|
197
217
|
sglang/srt/layers/attention/merge_state.py,sha256=OnWmNV6LYTUj0y8SCy_IK7A9q6v9bZwR4_XiLOIvbrk,1423
|
198
|
-
sglang/srt/layers/attention/
|
199
|
-
sglang/srt/layers/attention/
|
200
|
-
sglang/srt/layers/attention/
|
201
|
-
sglang/srt/layers/attention/
|
202
|
-
sglang/srt/layers/attention/
|
218
|
+
sglang/srt/layers/attention/nsa_backend.py,sha256=yyN3TRmPyxDBaYKsV-c5upK_wEVV3MyjJocw20DCAVo,33190
|
219
|
+
sglang/srt/layers/attention/tbo_backend.py,sha256=MyEQ29I41LCOHXm4_g7J7sXSs5bNKWRlSa2EfqploD4,9185
|
220
|
+
sglang/srt/layers/attention/torch_flex_backend.py,sha256=MnnZLxi41R-d4P0cyklNyefSWOU-AdFEjr8pxJWNgdI,11711
|
221
|
+
sglang/srt/layers/attention/torch_native_backend.py,sha256=NdWD2ArmS1lQN9mZIbS2rcHueBWColNxvhcLaQGAKGQ,9728
|
222
|
+
sglang/srt/layers/attention/triton_backend.py,sha256=3nIh_oQXVPDZw07m-bkXkAAddW39exOxzNMCDQbc5YA,43564
|
223
|
+
sglang/srt/layers/attention/trtllm_mha_backend.py,sha256=l6ZbRIZ9Q4rHbsULJ49_MKd32vWdr3IRF5Pi8Dnzv0M,27503
|
224
|
+
sglang/srt/layers/attention/trtllm_mla_backend.py,sha256=k3e-tD-YjiVs8kOzEXaakCI7g6MDLD99NXMtAscrBZY,29396
|
203
225
|
sglang/srt/layers/attention/utils.py,sha256=lx_jNw2VwlBe2VnDb1eXRRdN9WArw9rdWHTHqndUEFk,3174
|
204
|
-
sglang/srt/layers/attention/vision.py,sha256=
|
226
|
+
sglang/srt/layers/attention/vision.py,sha256=F_dBh1DpepBt97JOhvQo0sI9tSKBU3w3LOxLT66wBQg,21780
|
205
227
|
sglang/srt/layers/attention/vision_utils.py,sha256=6hPPAt8p3SOm7tgHNUesnMLxvQiKbaOM5EK6K-ig5dI,2771
|
206
|
-
sglang/srt/layers/attention/wave_backend.py,sha256=
|
228
|
+
sglang/srt/layers/attention/wave_backend.py,sha256=dMsR7GsfDYgVizY8QJ94ASrxiiJ5_MNOuTUDDQzvpPQ,22915
|
229
|
+
sglang/srt/layers/attention/fla/chunk.py,sha256=N8ToAspPlgR7sVJBImzpmlGV6rWsMA7aTaTa1iddcPg,8929
|
230
|
+
sglang/srt/layers/attention/fla/chunk_delta_h.py,sha256=Gf8RgLesGaR9GM4HTB7kDDCVirwR9XvO7O3I1HWdcuY,10681
|
231
|
+
sglang/srt/layers/attention/fla/chunk_o.py,sha256=IZllM88Ap1ztEF9nOSfcx7KmRvmwIwiBMqwT9AjjH-U,4980
|
232
|
+
sglang/srt/layers/attention/fla/chunk_scaled_dot_kkt.py,sha256=445twkm9l0Mu9Q_VFBIgUZQzTNLvc82gqIAe8ylip-4,4386
|
233
|
+
sglang/srt/layers/attention/fla/cumsum.py,sha256=beIT2WTSn8Wk_wknoWM3JYQhzmzR1JJwMFzspsXvWng,8374
|
234
|
+
sglang/srt/layers/attention/fla/fused_recurrent.py,sha256=yOSTwqcNSoPWwQD--tcCkVd2OwB5Ox0nLGnLLv8MRr8,20856
|
235
|
+
sglang/srt/layers/attention/fla/fused_sigmoid_gating_recurrent.py,sha256=dlmOu-7DG6YoSNexT3YV8GudrFPn_RiYj-3hClWdy-c,6424
|
236
|
+
sglang/srt/layers/attention/fla/index.py,sha256=kQGcR1doARV324QuA-faNoyo6u8a_wb-DfiPe_bOdhY,1050
|
237
|
+
sglang/srt/layers/attention/fla/l2norm.py,sha256=XOs5nGdWROp3FW4fHSObC8sa_x697yDV0jt5UKuLllE,3783
|
238
|
+
sglang/srt/layers/attention/fla/layernorm_gated.py,sha256=gNAAUyRfnjf5vTx92Gkah5UeumrlgH7oCq0Tg5D_5mQ,9938
|
239
|
+
sglang/srt/layers/attention/fla/op.py,sha256=Qc-zHwZY9-8Aw53AuuzNfpbPBwKgIvVMgtWcfJLl6A4,1695
|
240
|
+
sglang/srt/layers/attention/fla/solve_tril.py,sha256=-Z51hFsLER_Lzn8h21WRUajbVOA1wuu5TSEvaC2AvK8,14405
|
241
|
+
sglang/srt/layers/attention/fla/utils.py,sha256=yBYjW-blAsq_uZxZ961Dub3cUouIvl4DCKV9UqkQSJw,10474
|
242
|
+
sglang/srt/layers/attention/fla/wy_fast.py,sha256=zf6_76vq7yYtn9Ir1Le4RVOKLaPNF_4bNST3ePuH0cU,4407
|
243
|
+
sglang/srt/layers/attention/mamba/causal_conv1d.py,sha256=w1sBOwpJeDKeYK83vPaaep5Vhq2J_qey9JUaQ2vjnjU,4474
|
244
|
+
sglang/srt/layers/attention/mamba/causal_conv1d_triton.py,sha256=M04m2D13KhSCvsC37vuslNpwbgS4HLaJwWzyWjqev2E,37539
|
245
|
+
sglang/srt/layers/attention/mamba/mamba.py,sha256=HRGM1fFe_MzVY-6rdkYHndsaSaQB3cOWUCow8bPQYFY,24243
|
246
|
+
sglang/srt/layers/attention/mamba/mamba_utils.py,sha256=3-iipUQzuHPUDlWYhse2XwtiPYgcblOsiOoozkIoOZs,2820
|
247
|
+
sglang/srt/layers/attention/mamba/ops/__init__.py,sha256=r-IpDGRIwTqGJzZoWNGyHqsqqv8oF19pNHEb7_TZR9M,98
|
248
|
+
sglang/srt/layers/attention/mamba/ops/layernorm_gated.py,sha256=hmuL7vQ77aWFENCDsJDB7Nki5DDdMTEiQywbaU4FC04,5229
|
249
|
+
sglang/srt/layers/attention/mamba/ops/mamba_ssm.py,sha256=aCFsiBdBDYt3D5HT0lxt6Rru9-gPYZspBM_WpTGgHh4,13680
|
250
|
+
sglang/srt/layers/attention/mamba/ops/ssd_bmm.py,sha256=xZEjoVv5zdUeu_36WcS0VO9Qkmob2BwGgJ1RqQmg6w4,8420
|
251
|
+
sglang/srt/layers/attention/mamba/ops/ssd_chunk_scan.py,sha256=gyph59WoRu7rEuzARSK6731b_gCKx326vZ9X-Is1qbs,20380
|
252
|
+
sglang/srt/layers/attention/mamba/ops/ssd_chunk_state.py,sha256=yaZ1Hpj3XNSvFA0wSXkEBBZZKgy0441VrHgLBcp8r6w,24392
|
253
|
+
sglang/srt/layers/attention/mamba/ops/ssd_combined.py,sha256=ivE3zqD31W01W7_RgpRG8Tfh11Co4mjbz2H6fWzGBP8,8909
|
254
|
+
sglang/srt/layers/attention/mamba/ops/ssd_state_passing.py,sha256=lCt7NzzUbyCcnfDB2ai_agVM_hoZvfrB1d-UbFO8WWs,10377
|
255
|
+
sglang/srt/layers/attention/npu_ops/mla_preprocess.py,sha256=n6iR86qvKTYY_mC1EMxMiyHQaXZ8pF9o2iNCV68BA0E,14540
|
256
|
+
sglang/srt/layers/attention/nsa/dequant_k_cache.py,sha256=ie1qLy7AAipFYS0EJFrunN8Mq97D9v-noWaJBYMiv0I,4973
|
257
|
+
sglang/srt/layers/attention/nsa/index_buf_accessor.py,sha256=jPYRgFfTdH_g83OUs9q3ncj2Y036ikBvflMjYEjrHGw,11481
|
258
|
+
sglang/srt/layers/attention/nsa/nsa_indexer.py,sha256=FgK_o31MRswHX8CQiX9jPp04z_P7GgBumC91v_pRJHM,26466
|
259
|
+
sglang/srt/layers/attention/nsa/quant_k_cache.py,sha256=FYl6gp16uGU3ftIHkU5vFAjfdPnoKvWOg4l7TBKkGI4,8675
|
260
|
+
sglang/srt/layers/attention/nsa/tilelang_kernel.py,sha256=Iifd9YYEnDGkRhjK0NH-pbyTiQX0kVpJynyGwZcnH8w,31554
|
261
|
+
sglang/srt/layers/attention/nsa/transform_index.py,sha256=0yurXIDt5EQ9k_P5g_aDXohdzvSiBQcQpCHBcNf0ZBA,4745
|
262
|
+
sglang/srt/layers/attention/nsa/utils.py,sha256=qiUetSApDNI3Qz4_3-NVh_-mknU8ZHgA9sejb6PQKfY,901
|
207
263
|
sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=JnOtqtPXEgn_T0R4gGCROx1szlAhu_G_ODNKCazz0lE,20554
|
208
264
|
sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=FbgrPO5GvEGEvojV7Ss0jX7384e9vuzX2goDCQuP1Zk,31124
|
209
265
|
sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=eYg4xQoH5gBkM7bjzpBIpTdcelvUK3BujJ-8eXRCm70,17108
|
210
266
|
sglang/srt/layers/attention/triton_ops/merge_state.py,sha256=v9nD01a5eTnkwZxMwERtrrRfC5rs6GxkYOpErkAPcYI,2877
|
211
267
|
sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=waZsmpKIp8rTgFSoM4QMabJuLaB3yW6ltOzAKJksBoE,6260
|
212
268
|
sglang/srt/layers/attention/triton_ops/rocm_mla_decode_rope.py,sha256=664WnAJ91EiCUZOcnVDfbTQf4uGJ4ZDZB1CbxpEUFZc,13866
|
213
|
-
sglang/srt/layers/attention/wave_ops/decode_attention.py,sha256=
|
214
|
-
sglang/srt/layers/attention/wave_ops/extend_attention.py,sha256=
|
269
|
+
sglang/srt/layers/attention/wave_ops/decode_attention.py,sha256=oJNBG57cySgxtmJe1MSEsZKOIQOXnH0iVAcFO6DUWvY,4732
|
270
|
+
sglang/srt/layers/attention/wave_ops/extend_attention.py,sha256=LyZQQIvFYfcIcObgy_tETK5JxRLgAkOYlF-KhGcGEoA,3877
|
215
271
|
sglang/srt/layers/attention/wave_ops/prefill_attention.py,sha256=viTUit0rxjVV5UahuwmVAhmL66YRFJD4g1YhH_oRBu8,2344
|
216
|
-
sglang/srt/layers/moe/__init__.py,sha256=
|
272
|
+
sglang/srt/layers/moe/__init__.py,sha256=P8OC-pUE5yeEr80QUDqSPwF9a9BAYPuaK7GNlQovsJw,863
|
217
273
|
sglang/srt/layers/moe/cutlass_moe.py,sha256=JKJED-4709ndP5AwhQ7Vi04GJjw5d9Xl_mWOsZPZ3U4,14298
|
218
274
|
sglang/srt/layers/moe/cutlass_moe_params.py,sha256=9NRCmgP_Ug3gGqCcpi-x-QRbLjCNpw8792gKXwZsbEU,6522
|
219
|
-
sglang/srt/layers/moe/cutlass_w4a8_moe.py,sha256=
|
220
|
-
sglang/srt/layers/moe/
|
275
|
+
sglang/srt/layers/moe/cutlass_w4a8_moe.py,sha256=psvH2RhUQkK4oyaSE7xe0EncCqLvNbd_NwR4XHxk79w,7017
|
276
|
+
sglang/srt/layers/moe/flashinfer_cutedsl_moe.py,sha256=4TS3YLrmxn5dEHYUKeY7aohBXYeAsU8UCx846NXFTw8,6197
|
277
|
+
sglang/srt/layers/moe/fused_moe_native.py,sha256=XzJPM9Z6YQ4ejrMrBFt3l1i7guBcdQTXo9-gT5H8AIA,3479
|
221
278
|
sglang/srt/layers/moe/rocm_moe_utils.py,sha256=07Z99bTV3B-b2Cbm-odhGpx2twxtnVpYvaDMBE1K3LM,4555
|
222
279
|
sglang/srt/layers/moe/router.py,sha256=eUNu_Uz5VB2FOZzZyYuZo5pokCVBS17_fcjHQbmvDSE,12181
|
223
|
-
sglang/srt/layers/moe/topk.py,sha256=
|
224
|
-
sglang/srt/layers/moe/utils.py,sha256=
|
280
|
+
sglang/srt/layers/moe/topk.py,sha256=HcvKwbew1W4NJ-IGCp2aietW3o3KQPRuC7siRrjavt4,32240
|
281
|
+
sglang/srt/layers/moe/utils.py,sha256=HPRJy9cYYPAV5jU4dPYidIpVuSyk1HgaJrePQyD21Fo,6361
|
225
282
|
sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
226
|
-
sglang/srt/layers/moe/ep_moe/kernels.py,sha256=
|
227
|
-
sglang/srt/layers/moe/ep_moe/layer.py,sha256=
|
283
|
+
sglang/srt/layers/moe/ep_moe/kernels.py,sha256=WfHn5Kt3NH6MzawvtycMSgo_WDqg1aJQXp1HuqgNAP4,46168
|
284
|
+
sglang/srt/layers/moe/ep_moe/layer.py,sha256=7Aekn8YmGXThshcbMY7dk4fMVNLsTQ1_MnDLWwn5S6E,35667
|
228
285
|
sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=tis0ZJmih7gKHCurbLtY_o-bY3K4MOzQLYLC3ftIOf0,977
|
229
|
-
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=
|
230
|
-
sglang/srt/layers/moe/fused_moe_triton/fused_moe_triton_config.py,sha256=
|
231
|
-
sglang/srt/layers/moe/fused_moe_triton/fused_moe_triton_kernels.py,sha256=
|
232
|
-
sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=
|
286
|
+
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=mKH6zTTuh_Z8Ceg6d1VWjGBI3iPBn1YfGO6udxIRAtU,23306
|
287
|
+
sglang/srt/layers/moe/fused_moe_triton/fused_moe_triton_config.py,sha256=1iX5ox2KT5sgLHrN6l4svsgtn1-q_6514Q_F1v83-iY,7747
|
288
|
+
sglang/srt/layers/moe/fused_moe_triton/fused_moe_triton_kernels.py,sha256=QOdl5uhZ8--aryXnKObdspKonGcF6Tc_U9o_lC_opo8,27760
|
289
|
+
sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=1FGnf9QRSEPYdOZzvu-DFuz2DYDY0u7TaW0IuDYm-i0,39690
|
233
290
|
sglang/srt/layers/moe/fused_moe_triton/moe_align_block_size.py,sha256=U93mxPPU2RP7d3QRSvoG3OCtUQIK8YhQfHf1ZGeREmY,3284
|
234
291
|
sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=Ai06BZ7uxMnk0nPWQelgvi1rV9Z72FetRo6p7E3rsYs,10986
|
235
292
|
"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
|
@@ -398,6 +455,7 @@ sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=Ai06BZ7uxMnk
|
|
398
455
|
"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
|
399
456
|
"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
|
400
457
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=qc7qnAk3BYcP5m_eT8IPDfnvFrng_kiRCS36_Uy1M2o,3260
|
458
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=JfLZHVC84ZsHVIdhJ2jGT9r2ZqGGxBqIZx4HRjFc4n4,3254
|
401
459
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=768,device_name=NVIDIA_H20.json",sha256=T2sNHVZCLbyuO68ZeudUzQMBsAQzs2-eo53I-EUIuWM,3233
|
402
460
|
"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
|
403
461
|
"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
|
@@ -415,58 +473,72 @@ sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py,sha256=Ai06BZ7uxMnk
|
|
415
473
|
"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
|
416
474
|
"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
|
417
475
|
"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
|
476
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=352,device_name=NVIDIA_RTX_5880_Ada_Generation,dtype=fp8_w8a8.json",sha256=L7acUMDmYZOeA-riEds1-r9Wsq2KBWJlG7X7rilWjfA,3254
|
418
477
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=384,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=6JYGN4imwzIrQAEgboSO93wWYBuguzTgBjfTedq2c_Q,3250
|
419
|
-
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=768,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=dPe_9-JH0wF8q7Mhec4WWSLSGRE2gj_AQT3dnR3FREI,3257
|
420
478
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=129,N=352,device_name=NVIDIA_B200,dtype=fp8_w8a8.json",sha256=ClPogINm5AsnjDqx6Hz_rHrmnnCdUPid-jYIdIgjZxs,3250
|
421
479
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=129,N=352,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8.json",sha256=RMYwqtB7hYeaCyuZ1feVnRhBGlN4jMcmNQX2hqUqBlk,3254
|
422
480
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=129,N=704,device_name=NVIDIA_B200,dtype=fp8_w8a8.json",sha256=W8C1GtP4K43SK9128U52DD5WWofvPleAJE4us2Qju1k,3251
|
423
481
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=161,N=384,device_name=NVIDIA_RTX_PRO_6000_Blackwell_Max-Q_Workstation_Edition,dtype=fp8_w8a8.json",sha256=RMYwqtB7hYeaCyuZ1feVnRhBGlN4jMcmNQX2hqUqBlk,3254
|
424
482
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=256,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=T6kCtCUytCxiW8JOLfEF3YgdCD9_mKccJ85IqT0b8LU,3253
|
483
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=256,N=256,device_name=NVIDIA_H800,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=M3W1OQU6RgKAQL2EoYivAIidtz1cvgoK6_6ntNfn_lY,3265
|
484
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=256,N=512,device_name=NVIDIA_H20.json",sha256=bSe-KUwiVT3yM9jQuXlDmxe_KUft7FUn5zOby8bpSkc,3233
|
425
485
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=257,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=RbJYTRfjP_3-ytfj53W_w2jFpZ9xwMaqoKW52pkgP3c,3257
|
426
486
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=257,N=64,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=QEk8bE-GieZltTiv3ig__ScM9Q_4K-c75UzIuDrGmPA,3232
|
427
487
|
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=384,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FEnjuvvky-cvW1iq8TPrfGOOH8y4rOEx4jlYZ4tqbDs,3253
|
428
|
-
sglang/srt/layers/moe/
|
429
|
-
sglang/srt/layers/moe/
|
430
|
-
sglang/srt/layers/moe/
|
431
|
-
sglang/srt/layers/moe/
|
432
|
-
sglang/srt/layers/moe/
|
433
|
-
sglang/srt/layers/moe/
|
488
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=JUzq6vJzOAVw3VOXzCjBpaLpf5OxPQmo2iJPJjzMm6o,3241
|
489
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H20-3e.json",sha256=arTXsLka4xWkCERNBayRHqfhwrtqdd_Yhy-SziVVTZY,3230
|
490
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H200.json",sha256=HRGJWqy1CC7ouRY6e_t3C832CM0BIlQ0Zo6nofzBf8k,3238
|
491
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H800,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=2PKJ4-mLxVWDgm81JtkjRpz1mKcQztGdmR84ItlMWfQ,3255
|
492
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=256,device_name=NVIDIA_B200.json",sha256=PbtJ0dJWYABYkjoN_6vvsSRgbGkNvGFNjQT38qid1-4,3238
|
493
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=256,device_name=NVIDIA_H20-3e.json",sha256=3oOQFsbOjejH_zQaMYlKUtQ6SnxIH5_f3UIaFfSlc8U,3236
|
494
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=256,device_name=NVIDIA_H200.json",sha256=mmiNUNFajlYSjFXL0DMykS6I-D1P8AQXdRBpeCARD-0,3243
|
495
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=64,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=FeRAKkFUl3iMlNBBrb_Yoe4OsWyvSbVt5-wmXEjLbFQ,3237
|
496
|
+
"sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=64,device_name=NVIDIA_H200.json",sha256=phmhkQNz8DcLHPqN5tjmJqOgbX0BhQx9cP82fp_jyIQ,3236
|
497
|
+
sglang/srt/layers/moe/moe_runner/__init__.py,sha256=Dc4PH6HfzQDjJWL7iuZ-0qh8oRu7OtnH6IEl675pYiU,172
|
498
|
+
sglang/srt/layers/moe/moe_runner/base.py,sha256=GMdFI74BGO5HDjtel5QWI7W4CFix0BHkw_HrjVswhcw,8905
|
499
|
+
sglang/srt/layers/moe/moe_runner/runner.py,sha256=7ZOqDu0r1GY-9ra1VICq4WmH1nJVSQoE5yBXuSUr2CI,2669
|
500
|
+
sglang/srt/layers/moe/moe_runner/triton.py,sha256=YaPqGIX-5y25nxmjF54_ASYbr9Dw4RYze2cNV4cUg-Q,14433
|
501
|
+
sglang/srt/layers/moe/token_dispatcher/__init__.py,sha256=I5I1ySCHbyAW5cFQy9mpuqkSWqLclLmi2AspCkoP8ig,993
|
502
|
+
sglang/srt/layers/moe/token_dispatcher/base.py,sha256=BqqwuSETP47vTaHkUoQWTEeZcAkg0n_Dgka-PkFM4uc,4029
|
503
|
+
sglang/srt/layers/moe/token_dispatcher/deepep.py,sha256=LtsaJ3bR-HBQL9ZWjgbw-HaP-hcAqAgrmpO-829PL94,26293
|
504
|
+
sglang/srt/layers/moe/token_dispatcher/standard.py,sha256=_OSudEOHA5qHqR5TJqnylQxUYyk7ZnhRMGOf7sMCyt0,1526
|
434
505
|
sglang/srt/layers/quantization/__init__.py,sha256=GmJvVSlkJXf2AmTfWBN6EBrOhhYlxPYNE_eAEXMZfjs,7813
|
435
|
-
sglang/srt/layers/quantization/awq.py,sha256
|
506
|
+
sglang/srt/layers/quantization/awq.py,sha256=YuAkweqSvQQGTGQvg4VpNXevSaJvT8CROpiNzmY_3tQ,26980
|
436
507
|
sglang/srt/layers/quantization/awq_triton.py,sha256=4XwJbRNlFmlvjbMGtWF1Wvk26Kf96162vNEmuI5Kars,11737
|
437
|
-
sglang/srt/layers/quantization/base_config.py,sha256=
|
438
|
-
sglang/srt/layers/quantization/blockwise_int8.py,sha256=
|
439
|
-
sglang/srt/layers/quantization/fp8.py,sha256=
|
508
|
+
sglang/srt/layers/quantization/base_config.py,sha256=ql3kwwnjVzZGh8BUcimMt8K6kFxQtVbR5WgZHXqkOqA,7245
|
509
|
+
sglang/srt/layers/quantization/blockwise_int8.py,sha256=pqSLo4gwazZ4ajdVzyouwh_Z5oVDbWq9wo8RnrMdY8o,13998
|
510
|
+
sglang/srt/layers/quantization/fp8.py,sha256=pMr42f5f4QDQcOYmo3g2VBv_Dh8li85KpnbKYa2oNyo,50739
|
440
511
|
sglang/srt/layers/quantization/fp8_kernel.py,sha256=Mgigb6GmF66oCrgjS97IvDDSacNYb0u-VYAf4FChhqM,54849
|
441
|
-
sglang/srt/layers/quantization/fp8_utils.py,sha256=
|
512
|
+
sglang/srt/layers/quantization/fp8_utils.py,sha256=sxsjlqL7sWXwbgsxwfzFSkojyN3w8KQh1R99JEWdBl0,29191
|
442
513
|
sglang/srt/layers/quantization/fpgemm_fp8.py,sha256=dq7rJz-QB9LovmZZLBMK8YJEOr05wjVrtWLmSTkXjYw,7004
|
443
|
-
sglang/srt/layers/quantization/gptq.py,sha256=
|
514
|
+
sglang/srt/layers/quantization/gptq.py,sha256=Qe6XfBWd9aOW7jkLmQR61AjmJRe5MJi6FJvGno1HH4k,39433
|
444
515
|
sglang/srt/layers/quantization/int8_kernel.py,sha256=GfRC9FOn9exNvK4QHbUeBj3Hhv32VcyGphapFPt5b84,12625
|
445
516
|
sglang/srt/layers/quantization/int8_utils.py,sha256=YK9CS-lb_n91kNCTKK5o5apYF31V2giDg5G5VKrpcUA,2356
|
446
517
|
sglang/srt/layers/quantization/kv_cache.py,sha256=_9pF5rwvB7ta6Gdc5YKVVGbNzYwqmhIx4TrX1-xnodQ,3261
|
447
518
|
sglang/srt/layers/quantization/marlin_utils.py,sha256=aM3GU7b7rHNZE-tZjRSrtOaPWzSaMcLxck4he8MlEVI,26344
|
448
519
|
sglang/srt/layers/quantization/marlin_utils_fp8.py,sha256=iSjq-ikPOTU5plrWNTW4MletKLySIrNl5NhQmY6FyAo,12508
|
449
|
-
sglang/srt/layers/quantization/modelopt_quant.py,sha256=
|
450
|
-
sglang/srt/layers/quantization/moe_wna16.py,sha256=
|
451
|
-
sglang/srt/layers/quantization/mxfp4.py,sha256=
|
520
|
+
sglang/srt/layers/quantization/modelopt_quant.py,sha256=qDCoZxsZnZmi6RM3LbIw61Zzm3UM_vM9nRvIUOwcmCA,59352
|
521
|
+
sglang/srt/layers/quantization/moe_wna16.py,sha256=uoRhxJ8W_U_VFpbn03YEp0aCjbzqquWwonjuN7KBucw,19117
|
522
|
+
sglang/srt/layers/quantization/mxfp4.py,sha256=mb_DeJgrmEBS2CREZ09WsQVetxLozq3a4HhxqMq2amY,32993
|
452
523
|
sglang/srt/layers/quantization/mxfp4_tensor.py,sha256=bGeCXlIOhmFZFGWp_lITkIRDEgGUaLpawDvMy1cTOmA,5389
|
453
524
|
sglang/srt/layers/quantization/petit.py,sha256=74Jn0eICq23v_b4rEctDaYTINtbqkdM6IXycT-_FbbI,8954
|
454
525
|
sglang/srt/layers/quantization/petit_utils.py,sha256=-gy4zMhqNoIA1R0n7-5C0efV54jHonCUgPDUUFjGsyM,3245
|
455
526
|
sglang/srt/layers/quantization/qoq.py,sha256=jM96uJfKz3vo1B6SSIrQXNT3vnD3UJrepK2eY-tSQU4,8139
|
456
527
|
sglang/srt/layers/quantization/rocm_mxfp4_utils.py,sha256=Z4pYWl6ExOXq2qfTHorW16hVbJ-Rk4G8Wrxn_x8boHE,327
|
457
|
-
sglang/srt/layers/quantization/unquant.py,sha256=
|
528
|
+
sglang/srt/layers/quantization/unquant.py,sha256=T8w0mPvtBM1JhGhle6NoDY9C4xfmPWHU_VEJZWSnUNY,16079
|
458
529
|
sglang/srt/layers/quantization/utils.py,sha256=d4eaS4-Z4q3GRgb4HDMr2EoWEdQCefrVdJufK6n_NQY,18509
|
459
|
-
sglang/srt/layers/quantization/w4afp8.py,sha256=
|
460
|
-
sglang/srt/layers/quantization/w8a8_fp8.py,sha256=
|
461
|
-
sglang/srt/layers/quantization/w8a8_int8.py,sha256=
|
530
|
+
sglang/srt/layers/quantization/w4afp8.py,sha256=uSVtBXYIKPMa_Li1HHyZEifdaNzVCr-lTiEwRnIexD0,12148
|
531
|
+
sglang/srt/layers/quantization/w8a8_fp8.py,sha256=ugLV4G4MDog-UuLqSnEasBqE3Hh6AqE5ZYI371ean90,10415
|
532
|
+
sglang/srt/layers/quantization/w8a8_int8.py,sha256=e5mrRf7AWfit-5dOn3Ul9nTLt6OJth6SgrVpEdnVdE8,37076
|
462
533
|
sglang/srt/layers/quantization/compressed_tensors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
463
|
-
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=
|
464
|
-
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=
|
534
|
+
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py,sha256=Jfvvu0xL9WR075I2Qx8prTfGp77hKwkDH--V3R5pBpE,25928
|
535
|
+
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py,sha256=tFw8PIF5r8yjdJFt2_3Ou72vUJvgLd3bXFOWdgsCeno,26732
|
465
536
|
sglang/srt/layers/quantization/compressed_tensors/utils.py,sha256=mnUmKWFQUnY8bVoFHUuNVwqsfS-cefeR-ofyaihCXcY,7621
|
466
|
-
sglang/srt/layers/quantization/compressed_tensors/schemes/__init__.py,sha256=
|
537
|
+
sglang/srt/layers/quantization/compressed_tensors/schemes/__init__.py,sha256=ipQ5M4UkFHVsuAm2u4yZJd86NHtOenyeeahBtiLBLaM,448
|
467
538
|
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_scheme.py,sha256=tdKJC8c3SX8T3z8JL-1YCsg4ftcv55Wxt0vZrYftpX8,1635
|
468
539
|
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a16_fp8.py,sha256=VvqVpAr66AlS5IPVuIR5Nu0ESH0oAoKo545jNart6lI,5493
|
469
|
-
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_fp8.py,sha256=
|
540
|
+
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_fp8.py,sha256=4cfB6pv7kt8Ek0e88vLyXymSW6VqbAUpIjNt_-KK9-M,6318
|
541
|
+
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_int8.py,sha256=6usNdpr8mL5QRlTK_pZQtGwgmSMNi0KdMfm57kq3lw0,6908
|
470
542
|
"sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json",sha256=RdHQxWXwXqvio31192vsLaKjEr4f_DjpMPKlarY1IAk,3251
|
471
543
|
"sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json",sha256=0vLaJgo5B9ti-XMFKJuvSoMGjsZQ-RhHSx4cC8Xji-U,3254
|
472
544
|
"sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=tkLjwLC_aVXhzuvo-2QHkojXZauPJsf3jNHFn1S7uRA,3244
|
@@ -621,91 +693,112 @@ sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a
|
|
621
693
|
"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
|
622
694
|
sglang/srt/layers/quantization/deep_gemm_wrapper/__init__.py,sha256=esJMd0Yuj68t6QYOpmIFuiWP2J2dxTMC4bRBNH0Xk6I,26
|
623
695
|
sglang/srt/layers/quantization/deep_gemm_wrapper/compile_utils.py,sha256=PI9r3PHCMK9EgpoFuBgR4jvokA5sBz8zyC47ps3wet4,8164
|
624
|
-
sglang/srt/layers/quantization/deep_gemm_wrapper/configurer.py,sha256=
|
625
|
-
sglang/srt/layers/quantization/deep_gemm_wrapper/entrypoint.py,sha256=
|
696
|
+
sglang/srt/layers/quantization/deep_gemm_wrapper/configurer.py,sha256=a1b2P8-5PRNwW1ykXR7jJF94aUF3u3t3P9ad_R1k2PI,549
|
697
|
+
sglang/srt/layers/quantization/deep_gemm_wrapper/entrypoint.py,sha256=qiUuQn-InDLYVNDPFOxde2XB-qlE0fol0HPTkY5KN8g,3197
|
626
698
|
sglang/srt/layers/quantization/quark/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
627
699
|
sglang/srt/layers/quantization/quark/quark.py,sha256=EqQvWzAc6nd_yl6UIMIwmozKif4I8fRNfBum040UUEQ,14613
|
628
|
-
sglang/srt/layers/quantization/quark/quark_moe.py,sha256=
|
700
|
+
sglang/srt/layers/quantization/quark/quark_moe.py,sha256=fWUAEvLoaiw_6oNC7TQW2pUuR2AeNI9AaGWNcUGDiSQ,7446
|
629
701
|
sglang/srt/layers/quantization/quark/utils.py,sha256=zZHRe9mjxGp16_OJ5Qy_UtnQgipl3ZRZUmruYvyEm_w,7445
|
630
702
|
sglang/srt/layers/quantization/quark/schemes/__init__.py,sha256=z1LHBPqkxAS6Wp70DWwESOSixWpIhIQwz1IHcWDdURQ,167
|
631
703
|
sglang/srt/layers/quantization/quark/schemes/quark_scheme.py,sha256=b8-Djo-UQAwgxSWViKDoEP5dOSesnEBPyozX-osB3WY,1478
|
632
704
|
sglang/srt/layers/quantization/quark/schemes/quark_w4a4_mxfp4.py,sha256=sW57dU7HoxQGVbzUhyhRirGWA8oVpENB4NTzO7URGgU,4198
|
633
|
-
sglang/srt/lora/layers.py,sha256=
|
634
|
-
sglang/srt/lora/lora.py,sha256=
|
705
|
+
sglang/srt/lora/layers.py,sha256=0XlVcBTT1InMsBoLfiOO7yOnlDrnpXD_08wKo84T4B4,11852
|
706
|
+
sglang/srt/lora/lora.py,sha256=FooB4hOW5uvZJrDrdqKwBExb-lX5Uw_Y1gjU9an8hzc,7669
|
635
707
|
sglang/srt/lora/lora_config.py,sha256=qDgMTx_69jyJUl29O5FxLzYa0BMhqYVXWXfyyVOvGm0,1684
|
636
|
-
sglang/srt/lora/lora_manager.py,sha256=
|
708
|
+
sglang/srt/lora/lora_manager.py,sha256=F5kO3A6NFNFfu0hFZcn4Yr6Hum6si_kH6eE3Aw_Wxjk,19247
|
637
709
|
sglang/srt/lora/lora_registry.py,sha256=osbau-sQZ7BuCeGGpBMFy4w40GO5FGqRgRwDyxxgNmg,8167
|
638
|
-
sglang/srt/lora/mem_pool.py,sha256=
|
639
|
-
sglang/srt/lora/utils.py,sha256=
|
640
|
-
sglang/srt/lora/backend/base_backend.py,sha256=
|
641
|
-
sglang/srt/lora/backend/
|
642
|
-
sglang/srt/lora/
|
710
|
+
sglang/srt/lora/mem_pool.py,sha256=qh-AA7HhJDKG_xF5-67WGhrV6ohMM4TQU5oY40LVulc,11485
|
711
|
+
sglang/srt/lora/utils.py,sha256=FeIb51m5EIcetP1Pgn8X17EfuE5tGipm5iurRg3RAZw,4424
|
712
|
+
sglang/srt/lora/backend/base_backend.py,sha256=kfkr-OxdPyUwMzi0HgjTNfX9bZC4z_ljt3srzRPgMfI,6139
|
713
|
+
sglang/srt/lora/backend/chunked_backend.py,sha256=x2VNkjLHmrOXwI67v_2LgjpCH8R5wuSmY5jdTWV4K7g,12701
|
714
|
+
sglang/srt/lora/backend/triton_backend.py,sha256=bKH8DPo6fg7JVZt73iFV5BHc8amEJkr4FopbCnJer94,6072
|
715
|
+
sglang/srt/lora/triton_ops/__init__.py,sha256=no7N7uWnmCvHHbNZ6PZEadXDujIjHGP2jp_3N9j6vHM,495
|
716
|
+
sglang/srt/lora/triton_ops/chunked_sgmv_expand.py,sha256=sW7vsjOvyoI_00KJLZ5g0JwjTY7-AGzPTpKo3tuYC6U,6812
|
717
|
+
sglang/srt/lora/triton_ops/chunked_sgmv_shrink.py,sha256=sF9PzxK7XN3-MVqiz-S0DyVU2u1ADDhgnvpBe3GTaGI,5411
|
643
718
|
sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=QURCYxHNR8Ls4SQtt3dvdgjvdDVhywI9tOzsK8SV9m8,5779
|
644
719
|
sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=z9BB84Xu_MjwCu5v9Yvji-FkIzfEjDySAFybc4WYv7A,6483
|
645
720
|
sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=Ksova04wgeGsFqGOXWqJtMYaHgyUYcx8VU42BZQOkVA,5129
|
646
721
|
sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=VqCAFvUtq_l-0RGIkx3W_fzD55QcW2FAcqpvSHOBFp4,5115
|
647
|
-
sglang/srt/managers/
|
722
|
+
sglang/srt/managers/async_dynamic_batch_tokenizer.py,sha256=btIWhI15a3QCLabdZZhrJVEWZ3zZTgldbv7pGjAPZqo,7236
|
723
|
+
sglang/srt/managers/cache_controller.py,sha256=G-2292PKRRUQR6Co3sFt0oGs7zk6bqJ6RldEH0JQMGA,29300
|
648
724
|
sglang/srt/managers/configure_logging.py,sha256=8sNXZ2z9pBWOwn-X3wyz013Ob8Nbm1zDxRkxoZjH-l4,1633
|
649
|
-
sglang/srt/managers/data_parallel_controller.py,sha256=
|
650
|
-
sglang/srt/managers/detokenizer_manager.py,sha256=
|
651
|
-
sglang/srt/managers/
|
652
|
-
sglang/srt/managers/
|
653
|
-
sglang/srt/managers/
|
654
|
-
sglang/srt/managers/
|
655
|
-
sglang/srt/managers/
|
656
|
-
sglang/srt/managers/
|
657
|
-
sglang/srt/managers/
|
658
|
-
sglang/srt/managers/
|
659
|
-
sglang/srt/managers/
|
660
|
-
sglang/srt/managers/
|
661
|
-
sglang/srt/managers/
|
725
|
+
sglang/srt/managers/data_parallel_controller.py,sha256=YalvihRIJwph7OdSHf24rBsWYvmLQzUnJiDL261fyMM,14880
|
726
|
+
sglang/srt/managers/detokenizer_manager.py,sha256=QhJDFju95xTw7epUXvBH0t2DfwU-sR13v4i4ZAcFyQo,11815
|
727
|
+
sglang/srt/managers/disagg_service.py,sha256=7VyRva5oniAPs8G-DYg2fv-mdjSxk9cLZPzl_TmB_2Q,1588
|
728
|
+
sglang/srt/managers/io_struct.py,sha256=ZkxZpLbdEJTl92GtmaGfG0TDuuNpcC6YrvOXG6X75hI,48748
|
729
|
+
sglang/srt/managers/mm_utils.py,sha256=b6n39SlbaFP2XdUvJYOfRQL1Gch8fP--VxpOGsRJ8tc,30917
|
730
|
+
sglang/srt/managers/multi_tokenizer_mixin.py,sha256=audn1iSNNVu3d0DamslN1e-KGvhRJ5d_pw68nucQOUU,21676
|
731
|
+
sglang/srt/managers/multimodal_processor.py,sha256=0mLkwu6CGRQtt3G2T1tS3PSsh6dpyivV-vAfIHfcMbQ,1764
|
732
|
+
sglang/srt/managers/overlap_utils.py,sha256=QHeF3rQPI_kgMngPfDpAO8m80ZKSXOxB2G6r4Znyi2U,1809
|
733
|
+
sglang/srt/managers/schedule_batch.py,sha256=qQVFsvoJev5vgoOkFS4Zo_3tjv-5Vg0zxdxQlFYWS1Y,81929
|
734
|
+
sglang/srt/managers/schedule_policy.py,sha256=tdb81t-_zILXSGe5E3zNqg7N0_pbLWa8lfV1V7lAGnY,28192
|
735
|
+
sglang/srt/managers/scheduler.py,sha256=9hjS_FeM7Qfy1EYqvCmdrkuANFGFhNXvIXibDmGlZHE,119572
|
736
|
+
sglang/srt/managers/scheduler_input_blocker.py,sha256=SYZ91JNTHtqiBHUpkk_MboAiBFyl5U7ardLcYDAQ0OU,3689
|
737
|
+
sglang/srt/managers/scheduler_metrics_mixin.py,sha256=tvDovTLhtf8svgqVXIyoAc1xuf4PrH0AuCb0SPMYrGg,13197
|
738
|
+
sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=-ObOi1TK7WBwbGOzB5usSKmnGYjAqMPsYXMP81hPC44,33753
|
739
|
+
sglang/srt/managers/scheduler_profiler_mixin.py,sha256=62dv-rLS3acUncFNJd6S0P4jglJ5XY3CFIauB08xuJ4,11405
|
662
740
|
sglang/srt/managers/scheduler_recv_skipper.py,sha256=8rVQCBjn9f26Th4XbeUS3fEsvo1GI029Me28wMjcKRY,1103
|
663
|
-
sglang/srt/managers/scheduler_update_weights_mixin.py,sha256=
|
741
|
+
sglang/srt/managers/scheduler_update_weights_mixin.py,sha256=K9gCx16TaoeRah4fSYpIw_sOzRKuNUw2AJa6__dt5Bk,6194
|
664
742
|
sglang/srt/managers/session_controller.py,sha256=FGADRGSkRmS76o_BQ2fivpgw0Q5Q-K0Ws3aezH41YSc,6041
|
665
743
|
sglang/srt/managers/template_manager.py,sha256=M4EshVv1OhKxUAGQPP4OntRkSkUVvG93457xsYfWnQk,11968
|
666
|
-
sglang/srt/managers/
|
667
|
-
sglang/srt/managers/
|
668
|
-
sglang/srt/managers/
|
669
|
-
sglang/srt/managers/
|
670
|
-
sglang/srt/
|
671
|
-
sglang/srt/mem_cache/
|
672
|
-
sglang/srt/mem_cache/
|
673
|
-
sglang/srt/mem_cache/
|
744
|
+
sglang/srt/managers/tokenizer_communicator_mixin.py,sha256=NshY8ibxS7EQQGSYEp1PAAli0UL0-J_XZFtpQr9BIS8,25048
|
745
|
+
sglang/srt/managers/tokenizer_manager.py,sha256=YVCCbNP9g_56neSBcLnQCN1vtcmAp3apd7XLUq-NCig,77943
|
746
|
+
sglang/srt/managers/tp_worker.py,sha256=Qprdf6QhbddbKiKtO6Huu6J9a6jpWB_BH5qXADHvkas,14145
|
747
|
+
sglang/srt/managers/tp_worker_overlap_thread.py,sha256=9EoJ5ZbolfC8atvs6i6AFz9L9xHGj6OseaSLPezkVX8,11858
|
748
|
+
sglang/srt/managers/utils.py,sha256=fE_RZl5q0sZl9EQZNLlX9VNOHINxh-TJLQoH26GW2Lo,4279
|
749
|
+
sglang/srt/mem_cache/allocator.py,sha256=O-te5k_xLVB7o_GhBWCvperdm0_hJAlsxmw1QD3d9o4,18038
|
750
|
+
sglang/srt/mem_cache/allocator_ascend.py,sha256=aGsh_VMI0eYoAy3uKCYbpAMFxyXVTyu4JmdeGDWLLbw,4997
|
751
|
+
sglang/srt/mem_cache/base_prefix_cache.py,sha256=Bv84HD6e_yYdtmIQemK1m1OUViO0PnuAvOS9tDpQOJM,2795
|
752
|
+
sglang/srt/mem_cache/chunk_cache.py,sha256=5_yPQMGDP6TGLSxrEhOrBcrpd_7Z15Qewh6DOC6WyA8,3487
|
753
|
+
sglang/srt/mem_cache/evict_policy.py,sha256=yX8hFNQ_pdzuN4rMjYcByhQZLs-lRZdD0qDLkPe2ht0,627
|
674
754
|
sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
|
675
|
-
sglang/srt/mem_cache/hicache_storage.py,sha256=
|
676
|
-
sglang/srt/mem_cache/hiradix_cache.py,sha256=
|
677
|
-
sglang/srt/mem_cache/
|
678
|
-
sglang/srt/mem_cache/
|
679
|
-
sglang/srt/mem_cache/memory_pool_host.py,sha256=NXEVa1XjA8sXwpklvvWYjb0k6-gUgdj1P4MHBO9ybfA,26425
|
755
|
+
sglang/srt/mem_cache/hicache_storage.py,sha256=5L57L_vzqYyNk5PRdKikxyaaMj1LkCkWD38Sd93IAe8,7958
|
756
|
+
sglang/srt/mem_cache/hiradix_cache.py,sha256=hSsz8k1foizh-wJUxDi9h0NQZNNmSQoH_YP4A4uLtms,35414
|
757
|
+
sglang/srt/mem_cache/memory_pool.py,sha256=FiJzy8V39Jt7mIkyMiaq8OEhhnwP_GggOUm8nLiY3Ws,57038
|
758
|
+
sglang/srt/mem_cache/memory_pool_host.py,sha256=T3lgq0dyjyvOGdx4nGSGZzm6fqkX9H0S0JzYaNGJJ7k,27659
|
680
759
|
sglang/srt/mem_cache/multimodal_cache.py,sha256=zPnQLQhBZ6zsUpCQPSoNkrB9EEvpoDQS4mU7c3sRWjE,2171
|
681
|
-
sglang/srt/mem_cache/radix_cache.py,sha256=
|
682
|
-
sglang/srt/mem_cache/radix_cache_cpp.py,sha256=
|
683
|
-
sglang/srt/mem_cache/swa_radix_cache.py,sha256=
|
760
|
+
sglang/srt/mem_cache/radix_cache.py,sha256=1K1MvZZOLZVrbDti3pRrI1-pHJMlxxB3Scz9d9XNMo4,26390
|
761
|
+
sglang/srt/mem_cache/radix_cache_cpp.py,sha256=0twLAAR2EYWyfRiUVsb__s01ITraNEF9bXBPZFvul1U,9454
|
762
|
+
sglang/srt/mem_cache/swa_radix_cache.py,sha256=Kf5HAfnwJl1mr86pT7Y03o4RBKtUXfI6udbii9xAirs,40138
|
684
763
|
sglang/srt/mem_cache/cpp_radix_tree/radix_tree.py,sha256=tQZpz-H6HxNAQe9mYqt6aX8mfDZyu_fbJTcCNK-Ns1M,7179
|
685
|
-
sglang/srt/mem_cache/storage/
|
764
|
+
sglang/srt/mem_cache/storage/__init__.py,sha256=HDzHFUVeOHlry7HlfL5tRd5HUA02N94p8uzA_Va8nlM,251
|
765
|
+
sglang/srt/mem_cache/storage/backend_factory.py,sha256=rN8OfnkROk1rFQpJeCkiXMucuI--Ez35h4physV6Ys4,7822
|
766
|
+
sglang/srt/mem_cache/storage/aibrix_kvcache/aibrix_kvcache_storage.py,sha256=kYqfXHdWwmL8WPZ9Z5RtwPzKgb0Vq-QOTE-6YXndd5U,5463
|
767
|
+
sglang/srt/mem_cache/storage/aibrix_kvcache/unit_test.py,sha256=X7-GU8X5-t9adyVuD09obHQUo8d_Rfu--7R0C631p_g,3718
|
768
|
+
sglang/srt/mem_cache/storage/eic/eic_storage.py,sha256=V2isvEuzCsEZd1XzYFlwIyLbOm0P6rs4QXVqCDjRcPs,28628
|
769
|
+
sglang/srt/mem_cache/storage/eic/test_unit.py,sha256=PfNPpL624t23tyVgtmS5F1pGiuWDMrC39ZD6ZbkN828,3607
|
770
|
+
sglang/srt/mem_cache/storage/hf3fs/hf3fs_client.py,sha256=le_iTE5HZAbQNToa5E_dq_3U5ztu60ej-58tmzRtndM,5245
|
771
|
+
sglang/srt/mem_cache/storage/hf3fs/hf3fs_usrbio_client.py,sha256=6oH4v_WF_pKH4FMSr252v6QPhBCnRHztHqzuS8j6_QI,5323
|
686
772
|
sglang/srt/mem_cache/storage/hf3fs/hf3fs_utils.cpp,sha256=dDYPtxmL6OFhEyv_GuTgzkTDpBjcD0EK8kphnn7eaGc,1133
|
687
773
|
sglang/srt/mem_cache/storage/hf3fs/mini_3fs_metadata_server.py,sha256=lJ9Jp2kOgQQaUihxlO9l-W8fYfOMa1h1y0WAtzzFGLM,17850
|
688
|
-
sglang/srt/mem_cache/storage/hf3fs/storage_hf3fs.py,sha256=
|
774
|
+
sglang/srt/mem_cache/storage/hf3fs/storage_hf3fs.py,sha256=IeEc09wmBI5iRnN3rUSyfBxlFeHc92hHcA96dGRgCgM,21351
|
689
775
|
sglang/srt/mem_cache/storage/hf3fs/test_hf3fs_utils.py,sha256=g2h0Woa2CK39pRTigHCVR_hq095dTaC70lCIAIs9THM,1089
|
690
|
-
sglang/srt/mem_cache/storage/
|
776
|
+
sglang/srt/mem_cache/storage/lmcache/lmc_radix_cache.py,sha256=AdujdsJ_3ODf11O0LpA59X6yWWJLPQELelxE4hKU5-U,9925
|
777
|
+
sglang/srt/mem_cache/storage/lmcache/unit_test.py,sha256=DCSZDMJAe-HehL9hWzsP759pck36Q-8ijxWZKfgXmA8,3590
|
778
|
+
sglang/srt/mem_cache/storage/mooncake_store/mooncake_store.py,sha256=OWHraCjil547NYf6SjTrB6dteznqJL8kPoCFA1DCgwc,16885
|
691
779
|
sglang/srt/mem_cache/storage/mooncake_store/test_mooncake_store.py,sha256=6TiqbzEIL71DC7wL1B-Q-SMMZRYsz08sg-dZYeMzmAc,4930
|
692
780
|
sglang/srt/mem_cache/storage/nixl/hicache_nixl.py,sha256=bau9Jsp50lpEiy_osgb-dYzeoUkPWM03x7Q8z2BpKaU,9920
|
693
781
|
sglang/srt/mem_cache/storage/nixl/nixl_utils.py,sha256=KXkwMrbWwIOUOASINnUo4HyheiMd-93smmwGLv0CacU,7463
|
694
782
|
sglang/srt/mem_cache/storage/nixl/test_hicache_nixl_storage.py,sha256=vENVvGAd3PqcQEFA812wPoSBYhRemLr8Sfh1LQa0INU,9609
|
695
|
-
sglang/srt/metrics/collector.py,sha256=
|
696
|
-
sglang/srt/metrics/func_timer.py,sha256=
|
697
|
-
sglang/srt/metrics/
|
698
|
-
sglang/srt/
|
699
|
-
sglang/srt/model_executor/
|
700
|
-
sglang/srt/model_executor/
|
701
|
-
sglang/srt/model_executor/
|
702
|
-
sglang/srt/
|
703
|
-
sglang/srt/
|
783
|
+
sglang/srt/metrics/collector.py,sha256=s1BpU0cLgG1nHWYH5Y0PT0NbqRVHIs_IWql-UxOEEYY,34078
|
784
|
+
sglang/srt/metrics/func_timer.py,sha256=6-PYCeItVhsET6lc0BcFQfd0Tc3_wvJ8VHdp3mlWz7Y,3306
|
785
|
+
sglang/srt/metrics/startup_func_log_and_timer.py,sha256=PRVg3Z2d5GBMFifZXR-VhzGcn9nmKTkqEU41VJKZmaI,4937
|
786
|
+
sglang/srt/metrics/utils.py,sha256=MGb3quRI39n4tgJS1e0UHI5TBX3D0MwsgiZFMW2EsSQ,1903
|
787
|
+
sglang/srt/model_executor/cpu_graph_runner.py,sha256=ypp5yIW2OhRX-Xmido5mWuzEFpd-ohlq2c2wEuSW71I,23415
|
788
|
+
sglang/srt/model_executor/cuda_graph_runner.py,sha256=GcD2FkWG6CJRO3A23HY3Cu24ZE2c2MSpCD1NbtoLLgY,34348
|
789
|
+
sglang/srt/model_executor/forward_batch_info.py,sha256=834WSjXwpIeHGTgldRaBJ0R-YIjr8bops4YMzTHi9t4,42236
|
790
|
+
sglang/srt/model_executor/model_runner.py,sha256=sTRCfbndxVaUXFOaBUy9u6VPOoFFZHvm0_RBKuBF2Y8,89531
|
791
|
+
sglang/srt/model_executor/npu_graph_runner.py,sha256=HstTsIQF2QHCPMQH_AxKOu1Ulykw8q8Cpx7Q-u9Kq8I,3673
|
792
|
+
sglang/srt/model_loader/__init__.py,sha256=_1Y5sUZGxEEG_bmvLbSgYGa__qO60_d0blGB8sIdOgo,1020
|
793
|
+
sglang/srt/model_loader/loader.py,sha256=MhrqysDMxBTAuwoLbYyCfLJOluk_KSZWPhgiD-rP2Cw,64789
|
794
|
+
sglang/srt/model_loader/remote_instance_weight_loader_utils.py,sha256=_HnQD1Ab-Wv9oATOmwJe6aQaQD_Gs_yG9EsDCGJQWfs,2881
|
704
795
|
sglang/srt/model_loader/utils.py,sha256=KrOpyd_eefZC3zFnhnwxIKNTx9G-MmA6jIK3NQbTW0E,5002
|
705
|
-
sglang/srt/model_loader/weight_utils.py,sha256=
|
796
|
+
sglang/srt/model_loader/weight_utils.py,sha256=QbeiJY8VXIM1iLdG8X_JAjuy37mZzAUQzhBY_HHsxMw,44383
|
797
|
+
sglang/srt/models/apertus.py,sha256=UL2Rd1TDDXPb2N4rWyymI3JThG7KlrGIyBuDIFCitPE,26002
|
706
798
|
sglang/srt/models/arcee.py,sha256=Q6fujRESEnljzs8ls3XGsPtWxzfKfUa0QAvAekYzeBY,18981
|
707
799
|
sglang/srt/models/baichuan.py,sha256=HbvlErnkCSK4pRQYCSDxMcrn-1DQyfiNoeDcnRrJas8,15807
|
708
|
-
sglang/srt/models/bailing_moe.py,sha256=
|
800
|
+
sglang/srt/models/bailing_moe.py,sha256=IbG4BCgll0f5LUPkKZttizPU2wbtV5E9-dciUmjZ-EM,37726
|
801
|
+
sglang/srt/models/bailing_moe_nextn.py,sha256=ntuaXP8PXc_X_oqmd4fp_FzpRmnajKoWQynXwjsEN5M,6000
|
709
802
|
sglang/srt/models/bert.py,sha256=ODJe8YfNRP-hHsomFWk4_QpcuiSsNfjzGf256EDS0Pc,15802
|
710
803
|
sglang/srt/models/chatglm.py,sha256=cajLN9caBl09e0TwOFkiTTKDqwlbmHo_yS-NCjdeQW8,13957
|
711
804
|
sglang/srt/models/clip.py,sha256=kOPSXnmgH_QKu2-ENn49HLDS9yf7Gm1HCOE_YvoNhlA,19858
|
@@ -713,28 +806,32 @@ sglang/srt/models/commandr.py,sha256=5Y_b3K0QY7D37nFGkyiGgY38RleRui_GJUYcHSuHUZo
|
|
713
806
|
sglang/srt/models/dbrx.py,sha256=117gwrB_o8VlACwDH5YlXkPfuS7t2Mh-nyJvpNpIYxs,15903
|
714
807
|
sglang/srt/models/deepseek.py,sha256=M7nyHcAbTIs8vc1g6u00oy22-pp704Xp4T3clBzu4xM,17460
|
715
808
|
sglang/srt/models/deepseek_janus_pro.py,sha256=td8xGs6ARfJ8AQCYwUhMOZoWigrAs7m3trF5-kXCqik,70418
|
716
|
-
sglang/srt/models/deepseek_nextn.py,sha256=
|
717
|
-
sglang/srt/models/deepseek_v2.py,sha256=
|
809
|
+
sglang/srt/models/deepseek_nextn.py,sha256=QRcl5mWGDbYblV1dmo8MxPVeEUQQ_UhXumB5FEgj4M4,6255
|
810
|
+
sglang/srt/models/deepseek_v2.py,sha256=K6QdBihJEF64gurObwIVfJJ2icJroO538D2LQvSmoBw,132581
|
718
811
|
sglang/srt/models/deepseek_vl2.py,sha256=7X3gI4DE7guUtNJvMLf06jO8MUHKa2Aan8evZg5hsXc,13061
|
719
|
-
sglang/srt/models/
|
812
|
+
sglang/srt/models/dots_ocr.py,sha256=Dr9O-VBY1aSOFoIzEKcqs_QKRipjs8OQTfKrO5fJd-I,6792
|
813
|
+
sglang/srt/models/dots_vlm.py,sha256=0tbeQHbJryUMVqEnWelawDabzhcdwSj3Bb9-nj4YOKs,7242
|
814
|
+
sglang/srt/models/dots_vlm_vit.py,sha256=K6QzcHx38XAq9fmNRJ-iApScZ9FoV0lkn3JYkRsxplk,12044
|
815
|
+
sglang/srt/models/ernie4.py,sha256=3z14KFdYLR8J-zsm9oO1uBUeLlWrlkHoFKEYvKKJLKM,16079
|
720
816
|
sglang/srt/models/ernie4_eagle.py,sha256=E7f-ygCcY8pIoyTMKFyVC2uRcG59-_RaQTO4gd5_uWI,7223
|
721
817
|
sglang/srt/models/exaone.py,sha256=TpO-rtCpEZ8Ua7hGFnS8l2oAYhY0Pij50grc9WQ2mvc,13576
|
818
|
+
sglang/srt/models/falcon_h1.py,sha256=vtFgy-P6Jc6VoUibjpLNDi7633rcw61Fdyo8RDabiUE,20825
|
722
819
|
sglang/srt/models/gemma.py,sha256=2atKwFN5wCsJCmBRmURX9vaCcFjxlFirL7xsnG5KPlI,14180
|
723
820
|
sglang/srt/models/gemma2.py,sha256=n2s3WLGVM3XDvOXwbi_h7tgeh5SDDv1VvifH680DmV0,16785
|
724
821
|
sglang/srt/models/gemma2_reward.py,sha256=V8U3_ADUHWPdOwvEe1jhGW-oJmBgL8t1TY3-67Ksv2A,2618
|
725
|
-
sglang/srt/models/gemma3_causal.py,sha256=
|
726
|
-
sglang/srt/models/gemma3_mm.py,sha256=
|
822
|
+
sglang/srt/models/gemma3_causal.py,sha256=o46BNbaiOeoOv2xVU6-lWZhrrjyGskdebve0_wviWfU,27338
|
823
|
+
sglang/srt/models/gemma3_mm.py,sha256=r4r7OcYqKE-ZpDBuPZCF4TnLmG10MCwXupKRDR3E76w,17195
|
727
824
|
sglang/srt/models/gemma3n_audio.py,sha256=isgKfjA5UieYawxU6medL2ssXlzYPqAbagDBnLcemC0,36405
|
728
825
|
sglang/srt/models/gemma3n_causal.py,sha256=kBDlpSWAg6ZDmUPe08Q7sj5CbdVD_0h07zwJw8sKu7I,36335
|
729
|
-
sglang/srt/models/gemma3n_mm.py,sha256=
|
826
|
+
sglang/srt/models/gemma3n_mm.py,sha256=94QLW3FfAcqhz0E-gti8rTizTL9ci6CjMTxYOv2vHac,20371
|
730
827
|
sglang/srt/models/glm4.py,sha256=dijzVTTgg8zyJ3gga95MKY0WwqwnuAaf-VIV93OFAmc,11433
|
731
|
-
sglang/srt/models/glm4_moe.py,sha256=
|
732
|
-
sglang/srt/models/glm4_moe_nextn.py,sha256=
|
733
|
-
sglang/srt/models/glm4v.py,sha256=
|
734
|
-
sglang/srt/models/glm4v_moe.py,sha256=
|
828
|
+
sglang/srt/models/glm4_moe.py,sha256=_H_ldIZCuR22kQJ-5wZ3D-NW8rg7Mi8crSzWdBGO_RE,42037
|
829
|
+
sglang/srt/models/glm4_moe_nextn.py,sha256=pE9G2xr_hLBu6PpvWlyTVZ1BP_Eax2bVqwFHLv2d7kY,5962
|
830
|
+
sglang/srt/models/glm4v.py,sha256=gn2a4Sb_txTKOguYIzIgw_3DQX8IG1vhJMulh2gfvPs,23761
|
831
|
+
sglang/srt/models/glm4v_moe.py,sha256=wXbO-C3Kx6CEKGbzcs7CfdgH5giLjgrDDjJ7Nq7_rpw,17396
|
735
832
|
sglang/srt/models/gpt2.py,sha256=kclhxEs8oJk1KCyhmAqo7rZqecVGGHYkc-a1WZi3aIk,9841
|
736
833
|
sglang/srt/models/gpt_bigcode.py,sha256=1D6bi8Zu760gCRZkvdLHFcg8kCkY35ARwQYaMDtYhl4,10307
|
737
|
-
sglang/srt/models/gpt_oss.py,sha256=
|
834
|
+
sglang/srt/models/gpt_oss.py,sha256=7cjyN1v2cxP9vZNrMqeDorX20U9_I4YcSpz_b4x0zj8,43720
|
738
835
|
sglang/srt/models/granite.py,sha256=8q92shxVPAp_cJDohJATffSGd7Z0Oi-vF5jpY6DlK4s,19840
|
739
836
|
sglang/srt/models/granitemoe.py,sha256=j1rgZ62CbBioECjUblDCw_NneDQgY_QJODsI0fqXVO8,13779
|
740
837
|
sglang/srt/models/grok.py,sha256=8KCR13LtdnhswJrIwgb0sdFW0OxCA8GQ0fbN8gb856E,40518
|
@@ -745,18 +842,18 @@ sglang/srt/models/internlm2_reward.py,sha256=ndfGmyqYZbVZ7C7rJ-v9oK3wa-EpoBGybS8
|
|
745
842
|
sglang/srt/models/interns1.py,sha256=E4SyJWocXZ5rAuAV8kj569OHo57r04W4hmq1CF6oiFs,12516
|
746
843
|
sglang/srt/models/internvl.py,sha256=iQm2wDtkcyjPYoV9oY_jQ3_7WvhP315eP_vZEiQX3uw,27062
|
747
844
|
sglang/srt/models/kimi_vl.py,sha256=XulRiufQzOSWZe8-bUl2OEdznvsVU9OBsEYipaX3YkM,12877
|
748
|
-
sglang/srt/models/kimi_vl_moonvit.py,sha256=
|
749
|
-
sglang/srt/models/llama.py,sha256=
|
750
|
-
sglang/srt/models/llama4.py,sha256=
|
845
|
+
sglang/srt/models/kimi_vl_moonvit.py,sha256=0X07KnCsp9m8ZxVpdMSDSMH_yDZXkToIjwDtoyGX0Mo,23894
|
846
|
+
sglang/srt/models/llama.py,sha256=ZQc5XgkAtgr4etZ8tVp1B_hTk0pz_IehkVlCoaI5BXo,28263
|
847
|
+
sglang/srt/models/llama4.py,sha256=i4-Vo5FVE9zf_6Y80dyft5VjPOQ-a-xWYfEV8s-ltKQ,19872
|
751
848
|
sglang/srt/models/llama_classification.py,sha256=4QWTFaUZIFKYZvEzs8bx8VkOZNIwdYCLrnwrdAw4QK0,3108
|
752
849
|
sglang/srt/models/llama_eagle.py,sha256=Ubh_jmtfiOnriwvHgQT0ZGID7JvYvdSi_QGHOIkEgPE,5028
|
753
|
-
sglang/srt/models/llama_eagle3.py,sha256=
|
850
|
+
sglang/srt/models/llama_eagle3.py,sha256=lutyS4UaPg5QussPu08En3iO5e1lhsSm4dBXTm4R_XI,9201
|
754
851
|
sglang/srt/models/llama_embedding.py,sha256=zq-_lNu35VBFc7eemiam0zdkGIE8fzrgk5OWYfirZnA,3254
|
755
852
|
sglang/srt/models/llama_reward.py,sha256=LF2nqMV5XOrljGjAwJg43mBv3z6Q040I2EYlgZeCp8k,4681
|
756
853
|
sglang/srt/models/llava.py,sha256=xzYip_BAwpzSIdZre43LZiyTpFISa0ZCLdO6LUSbaCg,37702
|
757
854
|
sglang/srt/models/llavavid.py,sha256=-CSk0RJ2MQeb81sh-RISeVJFaI-XWY6nR6_I594MkME,12818
|
758
|
-
sglang/srt/models/longcat_flash.py,sha256=
|
759
|
-
sglang/srt/models/longcat_flash_nextn.py,sha256=
|
855
|
+
sglang/srt/models/longcat_flash.py,sha256=g_WX1UQgvkAtGGkw0Zk5NmDU9ew35zWKXn50CptKsh4,41016
|
856
|
+
sglang/srt/models/longcat_flash_nextn.py,sha256=MdXz-KDjvNexD1rT3tSjVCgAe1p0G24HKkO613fIG2k,29605
|
760
857
|
sglang/srt/models/mimo.py,sha256=Mp-iFp4YHuiuq-H8enUF5K5QbMnVcvEa6mURH6vM3yM,6140
|
761
858
|
sglang/srt/models/mimo_mtp.py,sha256=jSmqJAXu7G3OO7jW1oa2suI4H_Yl7u5ZT7w4lHFbHhg,7292
|
762
859
|
sglang/srt/models/minicpm.py,sha256=CzBJyZtfMpp8jvlEl29DHI7HLVq-CxuqP1UHwxIbaUI,14567
|
@@ -767,11 +864,12 @@ sglang/srt/models/mistral.py,sha256=ni7PppA_5X4ivTnIps3yoEaMEL-G6Gaf9b5hGC1vU5Q,
|
|
767
864
|
sglang/srt/models/mixtral.py,sha256=ohZVFJ75QWuE2kBKD7Sd4fcx9YbbjA7JxnYAsOIfORU,17166
|
768
865
|
sglang/srt/models/mixtral_quant.py,sha256=UUIq7lpUonD9IoGePyl59oJYEcVwT9wUXvtVp5tjQ9Y,15406
|
769
866
|
sglang/srt/models/mllama.py,sha256=OdX7brhyfGiMFPulxF77QOUi-Kcscg3Qo4ipyxzs9yA,39670
|
770
|
-
sglang/srt/models/mllama4.py,sha256=
|
867
|
+
sglang/srt/models/mllama4.py,sha256=ux93c-eLcG9LuL91f0euYHcMtYUt64VYmWXOo4JRzHQ,36255
|
771
868
|
sglang/srt/models/nemotron_nas.py,sha256=CoSAKqlbej4Xe30uY4w76iASBqeuJf8-imisDK00zEU,15964
|
772
869
|
sglang/srt/models/olmo.py,sha256=7-q_fA6XXdG7kPUjpUzYkzMUWJobuSjhqjYw9xSUs_c,12671
|
773
870
|
sglang/srt/models/olmo2.py,sha256=azmljhJF4ivcQfUtfsAUxq3ducE4tRKTL6iwe0IKYMg,14327
|
774
871
|
sglang/srt/models/olmoe.py,sha256=zryYUQEhEdhd6xcR_HyLV7gHCcKt-HWHeEeIUqmhmcc,16100
|
872
|
+
sglang/srt/models/opt.py,sha256=1fdsQU7yXSN9HEWb0ZVdiAF8wLzAunW4XomIrlHyu2o,23476
|
775
873
|
sglang/srt/models/persimmon.py,sha256=86Duo3HiZqDj9O2UzWbJgUEhQOXzGBUdhUofQhcs1VE,11244
|
776
874
|
sglang/srt/models/phi.py,sha256=n6A1d47MvnR1lmSsZ3TuzXk5kT1Lntj_YQ5uNvR0s5o,10287
|
777
875
|
sglang/srt/models/phi3_small.py,sha256=eJb8aS_3KLObrq3PNkoIyVHaQ6SZTAJe42rbpC54QKg,16012
|
@@ -782,35 +880,44 @@ sglang/srt/models/phimoe.py,sha256=EGaKFi02MZP20qFrILA4o2SJ0wb1HoJagFTSqVUbJcQ,1
|
|
782
880
|
sglang/srt/models/pixtral.py,sha256=ZkZrV0mEfmbBdgRko2AvsMFZbAVskgp_5a5p1xCSRN0,16835
|
783
881
|
sglang/srt/models/qwen.py,sha256=j0OU3JjznWi_TbayI09yYGM-Obov6mQzJO82Hs0qBGw,11868
|
784
882
|
sglang/srt/models/qwen2.py,sha256=kHNZCtT0jnQXC2aJN7Bp1DzaEzKWuhUAGL3hAmt8OU8,23598
|
785
|
-
sglang/srt/models/qwen2_5_vl.py,sha256=
|
786
|
-
sglang/srt/models/qwen2_audio.py,sha256=
|
883
|
+
sglang/srt/models/qwen2_5_vl.py,sha256=rMmmoc1ba3GjIE5bfcaEBSh62hNuBZT0tlEpiY-9N8Y,25031
|
884
|
+
sglang/srt/models/qwen2_audio.py,sha256=Q6zUfBdUTXDa7qx_aDVw6et7wonJsvBbrABmxLa_1YA,7782
|
787
885
|
sglang/srt/models/qwen2_classification.py,sha256=dGrMm4ebd30_lBhHOhaV57ig2iOTx3nqB4GEzsrRIM8,2747
|
788
886
|
sglang/srt/models/qwen2_eagle.py,sha256=6spFg6JYQGeUXLv2Q5eMkVzZvnqGyvW6SdVBtFMxlTM,4806
|
789
|
-
sglang/srt/models/qwen2_moe.py,sha256=
|
887
|
+
sglang/srt/models/qwen2_moe.py,sha256=LfqgHflkQFhiRVSuRnKBaCHW-zf4HKRN7q5ZGm6bGl8,30854
|
790
888
|
sglang/srt/models/qwen2_rm.py,sha256=-mQXDEv11p-I1HXgYLTtY6ROem6UYorO958WsDrzsgs,2837
|
791
|
-
sglang/srt/models/qwen2_vl.py,sha256=
|
792
|
-
sglang/srt/models/qwen3.py,sha256=
|
889
|
+
sglang/srt/models/qwen2_vl.py,sha256=HwNc08L3zgVfpD59QFcyrDefbems_dfLgapJpTBRhRw,21372
|
890
|
+
sglang/srt/models/qwen3.py,sha256=2OSKkk-AdIpg6ZnNcQ9tqZLlF-VH-QxncY1J_g8k-bc,19637
|
793
891
|
sglang/srt/models/qwen3_classification.py,sha256=bdjAylXJCjXKjAho74nmtIMpEG0Af98N-aZQ-pKKMY0,3262
|
794
|
-
sglang/srt/models/qwen3_moe.py,sha256
|
795
|
-
sglang/srt/models/
|
892
|
+
sglang/srt/models/qwen3_moe.py,sha256=AcOGpft3b-p31LII8jA2sRDK497fIicxFNIDGQbKG28,34083
|
893
|
+
sglang/srt/models/qwen3_next.py,sha256=nvQ2VVUNjHvU5cxXgx5Bh5EWMtkPMfHPYP9Ni45J3hI,37364
|
894
|
+
sglang/srt/models/qwen3_next_mtp.py,sha256=F-NnFGer8KxijNKIZoKICDfp6iFwct-XV0-q0KxAApg,4465
|
895
|
+
sglang/srt/models/qwen3_vl.py,sha256=2kFQAZ9yq9GNj7aVF9ktyghsTFtNOyHczBYLNF2X4UQ,29368
|
896
|
+
sglang/srt/models/qwen3_vl_moe.py,sha256=Bvx7dqzg0XzipiYqioCtVkwqxGSh8Y1zyvbMQ6D7HL0,18625
|
897
|
+
sglang/srt/models/registry.py,sha256=BpYy-E2m5jPmCNmue_o3G9lCQza1ge7ftmTJQ78pFGc,4276
|
796
898
|
sglang/srt/models/roberta.py,sha256=3k53V2Gbezk3jU-D03thXx8csGn5DxFK01ZQ9WFiWPI,9828
|
899
|
+
sglang/srt/models/sarashina2_vision.py,sha256=P71n-U3XqUcavrWNBAjpQ2V4P1fkZJOKMOqrIvfbNSg,10945
|
797
900
|
sglang/srt/models/siglip.py,sha256=DIVJRwdtpLD2QT1kVPIHw7Bn9BE40xJbkHujDNzkjgA,9892
|
901
|
+
sglang/srt/models/solar.py,sha256=BM30Eyn5Ptc0PCA6Lx3Hn7LZ_fV05n0rVm0vY_ILv5Y,18347
|
798
902
|
sglang/srt/models/stablelm.py,sha256=0x_31uIr3WcWwecdPAI3ek9KkyKBJS7VwknTk2y0gjY,12281
|
799
|
-
sglang/srt/models/
|
800
|
-
sglang/srt/models/
|
903
|
+
sglang/srt/models/starcoder2.py,sha256=ts6LDIjQ2FVnyLk4s2K9COD01RRhewJDW-VlvoyAk24,12377
|
904
|
+
sglang/srt/models/step3_vl.py,sha256=4j7drPJT2h-QwFH7tK7zR96xnuGvrxNbestuZsflCxg,35666
|
905
|
+
sglang/srt/models/torch_native_llama.py,sha256=3xf6CzCiizJ8i94GIChpOLq9q64Xa519RtuEspPTZQk,18501
|
801
906
|
sglang/srt/models/transformers.py,sha256=laA2h39MPeiAWKv4DEg0ydre-1pyXWFSRga-nwy1HUk,10244
|
907
|
+
sglang/srt/models/utils.py,sha256=NNKR-N_BGXRxbO6hCSwcq9kF9gZXCOfFc_8PPyo_aFs,1748
|
802
908
|
sglang/srt/models/vila.py,sha256=3NiRHWOqXnCRjXmswAldsNtiE5zRILbScEsbhUzLZF4,9846
|
803
909
|
sglang/srt/models/xverse.py,sha256=DsNVI9JpzN4jj0Ry6aTrj7r-xq5YLOoDX2kH4YLJA-I,14035
|
804
910
|
sglang/srt/models/xverse_moe.py,sha256=nBtG_PAK8wcp1tW2l2Zby2oEOmtWAYzH5COH-_N5IAM,17072
|
805
911
|
sglang/srt/models/yivl.py,sha256=oToK7-u5IGO7xwpJIQ7VtudlK6-zPqJX4bt6_wv0SH8,4850
|
806
912
|
sglang/srt/multimodal/mm_utils.py,sha256=6bV1_a7Iet9W0y3qdd0E-hFM78zalIzNZCYbLzJX2y8,12640
|
807
|
-
sglang/srt/multimodal/processors/base_processor.py,sha256=
|
913
|
+
sglang/srt/multimodal/processors/base_processor.py,sha256=1u6ewRgHL-svE1JSZ95iJVDFZIslfvBZne_LDR8YMPc,24902
|
808
914
|
sglang/srt/multimodal/processors/clip.py,sha256=5nXZuQYbyf-5mlqPwjhQrlWqxvswImWZQXivChsiHfQ,1057
|
809
915
|
sglang/srt/multimodal/processors/deepseek_vl_v2.py,sha256=Pymag9f0U5L_bBOTFs2q94BbuPJ-VwqSixs3meO0nsM,2569
|
916
|
+
sglang/srt/multimodal/processors/dots_vlm.py,sha256=1Lk8eE7y4g1Sdn_1CfR4EOI4deVHeJRlgHLmS7UUNBw,3533
|
810
917
|
sglang/srt/multimodal/processors/gemma3.py,sha256=7aD6cR60iSC_MqjzdCDMAgtrb_cmy9YJyqLgVRClBT8,1986
|
811
918
|
sglang/srt/multimodal/processors/gemma3n.py,sha256=mtmvBasOMxb3Y2NLZR02nMYX30RLUXhDYJ5uVOElr10,2776
|
812
|
-
sglang/srt/multimodal/processors/glm4v.py,sha256=
|
813
|
-
sglang/srt/multimodal/processors/internvl.py,sha256=
|
919
|
+
sglang/srt/multimodal/processors/glm4v.py,sha256=SQXatu6t-YgE8wNSavBRliN7J4DqiP57lmA47-zRlYQ,4953
|
920
|
+
sglang/srt/multimodal/processors/internvl.py,sha256=Po3_Dv71vKJidyyTY-KJRzGtxU2mCYIsETtcg9GhDg0,9853
|
814
921
|
sglang/srt/multimodal/processors/janus_pro.py,sha256=_4ltWmdtFjbh1VCMRbcgxW61eWIABIzHG-YgwHWaYq8,1471
|
815
922
|
sglang/srt/multimodal/processors/kimi_vl.py,sha256=5i8KmGxnnDtQc7XqUBNyanAfGxhxj8iRFBFONHmpiFg,1565
|
816
923
|
sglang/srt/multimodal/processors/llava.py,sha256=-YtDc_7TsMWmwKkD4KJZW2Y0jPFlZtbQ38K_N24hk0w,8231
|
@@ -820,66 +927,89 @@ sglang/srt/multimodal/processors/mllama4.py,sha256=1YGETKSzMZIa7BOwq3T9cbsYjUcW5
|
|
820
927
|
sglang/srt/multimodal/processors/phi4mm.py,sha256=8GaUVpGucV4UKoa0XkxN0gDpMZax6UyCiRBX_QaJK78,3857
|
821
928
|
sglang/srt/multimodal/processors/pixtral.py,sha256=Lx5Z-9NkovLDTHlRBznHxGn4r2XhDz8am2XpOskOVQY,3203
|
822
929
|
sglang/srt/multimodal/processors/qwen_audio.py,sha256=xkIFrHepF8Ok6MB-1OwVz1BmzPmEPgUe8lIb0azO2Bc,2411
|
823
|
-
sglang/srt/multimodal/processors/qwen_vl.py,sha256=
|
930
|
+
sglang/srt/multimodal/processors/qwen_vl.py,sha256=3ZjxgINjukJ953Vj6c67v8_x0Kh_9xq-1jOgVO-mqa8,10983
|
931
|
+
sglang/srt/multimodal/processors/sarashina2_vision.py,sha256=E0KSOw_yzc7prHkQfMHS_DzOcYL1GZJ77NopVkd_V5M,3024
|
824
932
|
sglang/srt/multimodal/processors/step3_vl.py,sha256=9OI6UW9ybGbAnYN_S-5jh7TN8BGM_bQkvh1AHBJKzAQ,18350
|
825
933
|
sglang/srt/multimodal/processors/vila.py,sha256=vWgV1lTdjiFiTR8w-7fC66-L_tpOOD59c-c9NljGelE,2192
|
826
934
|
sglang/srt/parser/code_completion_parser.py,sha256=KFa95OU0TeVjJkOqIgS4xV3kaJ5dFWOmAAgISyc1oEc,3803
|
827
935
|
sglang/srt/parser/conversation.py,sha256=cmVT4SCqRYxprYvmftUC_z_TkhXCqYC9B2rVZsmQe30,38869
|
828
936
|
sglang/srt/parser/harmony_parser.py,sha256=YweG4azWm5jmzYGaAHEXSaORpgWctPpSGQk_jp5H-CE,23227
|
829
|
-
sglang/srt/parser/jinja_template_utils.py,sha256=
|
937
|
+
sglang/srt/parser/jinja_template_utils.py,sha256=_rtLpriMmodo-tBD8B9ifu3BsZFdaQQCDPRUqF26pIg,8006
|
830
938
|
sglang/srt/parser/reasoning_parser.py,sha256=3P5viZ4oCsz695rZULWnRKSYBJxouHjihdpZMBuY9yc,11159
|
831
939
|
sglang/srt/sampling/custom_logit_processor.py,sha256=AwODYVJdRkcQ8PGtJrhzKsqAgn8XZLQbAmR9fGiQzmc,1608
|
832
|
-
sglang/srt/sampling/sampling_batch_info.py,sha256=
|
833
|
-
sglang/srt/sampling/sampling_params.py,sha256=
|
940
|
+
sglang/srt/sampling/sampling_batch_info.py,sha256=a-t8Al8D4x4DRqTnvmnTA4jHzOVp6JEFNJxUat7mDqE,15498
|
941
|
+
sglang/srt/sampling/sampling_params.py,sha256=yRkiwoZz0OrmmdMn2v63O5qkbns3Gv9MonxKYvyYcKc,6690
|
834
942
|
sglang/srt/sampling/penaltylib/__init__.py,sha256=mtN8grFEcaBUhl4yBHmw8NNirt_i6uKO2cDNLHOpZQE,496
|
835
943
|
sglang/srt/sampling/penaltylib/frequency_penalty.py,sha256=Loc3qjJTksNc5s-DV7QZHjgqoo5pxk7-nZzxwyhD2tQ,2144
|
836
944
|
sglang/srt/sampling/penaltylib/min_new_tokens.py,sha256=rdU_D7RoIcrQPhysNQEzmr4TO2OoEi___p-i3QdwkgU,3331
|
837
945
|
sglang/srt/sampling/penaltylib/orchestrator.py,sha256=Romk41ahToaVlNMnnd1Enko7_kRA4KdMGO94qnEbDHA,6061
|
838
946
|
sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=NRh10AJrrQlGJ6S-enGdRefrTrWpyqrSm-aNnyqQNQQ,2119
|
839
947
|
sglang/srt/speculative/build_eagle_tree.py,sha256=O9LJNaBflgJdWT94D-rGH1gJFJ18nst2oOD8HnA2mZ4,12859
|
840
|
-
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=
|
841
|
-
sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py,sha256=
|
842
|
-
sglang/srt/speculative/
|
843
|
-
sglang/srt/speculative/eagle_worker.py,sha256=
|
844
|
-
sglang/srt/speculative/
|
948
|
+
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=7QpNIGrZE8GEqe_B1vuLhpWPCAV9jY0Kr67Imbirycs,14201
|
949
|
+
sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py,sha256=nqnlTLzqfz-zYJvIFALXzfgbCJdrtslIhfCnxEkvQ5I,15897
|
950
|
+
sglang/srt/speculative/eagle_info.py,sha256=3JtRb60iCQzzi1JgreRhpGzC1aTs5Zbqv8XBDknnGHQ,29942
|
951
|
+
sglang/srt/speculative/eagle_worker.py,sha256=bAvFoMs1Mc5YcwEQhfysuaKIPvZOLdb6QqUeqNSrrFA,46909
|
952
|
+
sglang/srt/speculative/ngram_utils.py,sha256=U3a-c2bhNeIqZV4sKbzh3Hz_lAiqN1YFuFrbiaf0nsU,16600
|
953
|
+
sglang/srt/speculative/ngram_worker.py,sha256=F4TDjVNiwzD-iQrqFArTybODtITdZLpFUYTtGdCPc2I,9727
|
954
|
+
sglang/srt/speculative/spec_info.py,sha256=R3FK3ghFBUDSIUuNUYf41kFwEzP2fDBIlGdc45sRc4Y,2608
|
955
|
+
sglang/srt/speculative/spec_utils.py,sha256=7l-jkBg-of_5NJJ1boH98GN4YyAbjxfs39gZzykfK_o,20072
|
956
|
+
sglang/srt/speculative/standalone_worker.py,sha256=jcM_-7DMtRUMC9R0YeD6815kaWuftkFe8LIo7ThLVCc,4069
|
957
|
+
sglang/srt/speculative/cpp_ngram/ngram.cpp,sha256=j8D4xNaa8dv8CN9F5TG_yHxil-A79XiuMjyEE3ZUOSs,12962
|
958
|
+
sglang/srt/speculative/cpp_ngram/ngram.h,sha256=9OMOzO3KXEVbT5gEXjK-IRATI63in54LjeT_Xba5DSU,2492
|
959
|
+
sglang/srt/speculative/cpp_ngram/ngram_cache.py,sha256=Y22AM1CC0zOOarMWJzZQKz1-IMk3d4isSGkk-Uti74I,4383
|
960
|
+
sglang/srt/speculative/cpp_ngram/ngram_cache_binding.cpp,sha256=yrRXulpjivoSveCRL5ORfZ0hj2yXqjdVqc88APUJllg,1911
|
961
|
+
sglang/srt/speculative/cpp_ngram/param.h,sha256=tLSD_Ke1c2GjTY9tl3p1UgpbrBfaZqPRsHfw2YJ87hw,4375
|
962
|
+
sglang/srt/speculative/cpp_ngram/queue.h,sha256=m8OK4Nd6_yWnJkWxx-JRqUIOfzQwi3apy_3TvyxSY4Q,1288
|
845
963
|
sglang/srt/tokenizer/tiktoken_tokenizer.py,sha256=KJwr6Xlc7AdHegMTnAVKJlPpW59Z7AE27GTzFL78GwI,6222
|
964
|
+
sglang/srt/tracing/trace.py,sha256=5SKy8XA4KLaTcBO6ZJvJSP_IYK5--2RJN0Q6O6Vg10g,16110
|
965
|
+
sglang/srt/utils/__init__.py,sha256=M47TWjBX488QHsDCOTKHv-979SGOsnSu1wIWEqUaXnI,86
|
966
|
+
sglang/srt/utils/common.py,sha256=gbj2-BL4ffLEkdQBqBs93NYU8QWwA3UN8qOEVZua2w0,108529
|
967
|
+
sglang/srt/utils/hf_transformers_utils.py,sha256=7nUwqdrlkJMbi4aB5Ebvcf5NDVkLTzX4VVTNZav9Nps,17263
|
968
|
+
sglang/srt/utils/patch_torch.py,sha256=yaVs6YNm63rFYBw-v59usYR3cmb_qX6F69L5dMfQCxg,3307
|
969
|
+
sglang/srt/utils/poll_based_barrier.py,sha256=9vjn32ROlSueEk7EuZjWKUrrL7XVR9KJsSVliGTYB_4,937
|
970
|
+
sglang/srt/utils/rpd_utils.py,sha256=F8FWL_mT05xKQYLrzCJ8Q_5zfZJ86dHxX6LEqpwC7rw,20743
|
971
|
+
sglang/srt/utils/slow_rank_detector.py,sha256=81-KXp33wevk4jG2QuThsc17fLdmdeMouxwVfSXscic,2052
|
846
972
|
sglang/srt/weight_sync/tensor_bucket.py,sha256=vp_LRrhULNnxiw9Tf56FcEckgUg07lifgv1lasG6y7o,3695
|
847
|
-
sglang/srt/weight_sync/utils.py,sha256=
|
973
|
+
sglang/srt/weight_sync/utils.py,sha256=1FHM5kBav1GJUyYQLKHSJAh9MXwsuizDWxAVJMg1ZgE,4155
|
848
974
|
sglang/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
849
975
|
sglang/test/doc_patch.py,sha256=P9PYjBIVlZHvP9M34qS-8m1scmDP27ciNJNvZOdrUKE,1614
|
850
976
|
sglang/test/few_shot_gsm8k.py,sha256=zalplyHOYCnIwAnSQg-xlZl-Q-wGuMfjLdl9EJnUN7U,4333
|
851
977
|
sglang/test/few_shot_gsm8k_engine.py,sha256=h88s3HzEStA1zjQkR-AgIEOQMwJMSnPHkX_wGBWXUYw,3907
|
852
|
-
sglang/test/
|
853
|
-
sglang/test/
|
978
|
+
sglang/test/get_logits_ut.py,sha256=XotfHBDbQDRuSrJLRErsrdIOA0WDflnrK9171E-2iWE,1822
|
979
|
+
sglang/test/run_eval.py,sha256=1WaGfgsxCQWrvaIWS-m4lWjfTW-Y7IcASKvTG0Q-9gM,6420
|
980
|
+
sglang/test/runners.py,sha256=Tn__rDchJRP7pAorTS6gAAtqdPIeexpoEvwOwohTsB8,34105
|
854
981
|
sglang/test/send_one.py,sha256=XNZdfZs21tXDBwrDj2pq46WQd5smiKoYETY8U09c1Z8,5079
|
855
|
-
sglang/test/simple_eval_common.py,sha256=
|
982
|
+
sglang/test/simple_eval_common.py,sha256=Fm9PYFk6bYCb8RyMWr-ODfoNMUIftahTTOCBh84qiM0,12888
|
856
983
|
sglang/test/simple_eval_gpqa.py,sha256=M6hpPs8l2Z4wUiyE7rCU1zyKoImXB4Z395Hgz43uvvQ,3293
|
857
984
|
sglang/test/simple_eval_humaneval.py,sha256=s8YuKIBkNUEOBP4XKsTMRdXGv5aRhl0oJyl-89lGk1w,5687
|
858
985
|
sglang/test/simple_eval_math.py,sha256=6kGKNwNbLN-Af3Wj8WTimWhH-Xp3enDmSvvSjsgWUpk,2550
|
859
986
|
sglang/test/simple_eval_mgsm.py,sha256=rd7TSUyxdKbrXaVoewo24V8lCo_6kO8zxPhhmvylpw8,10259
|
860
987
|
sglang/test/simple_eval_mmlu.py,sha256=FkwamjGMjueTixymkedF-YiPloSLiy4ftILFUrKZ9XI,4357
|
988
|
+
sglang/test/simple_eval_mmmu_vlm.py,sha256=XpADoB825LUX-n4hDFOR3wkuowdLPB6sMHJEvM2dah8,15195
|
861
989
|
sglang/test/test_activation.py,sha256=bDBFl4Y30siXj7Ifl66fc2HytOrcZdHFiP2i9K8TIsY,3140
|
862
|
-
sglang/test/test_block_fp8.py,sha256=
|
990
|
+
sglang/test/test_block_fp8.py,sha256=CNQE3CildtLO49ZegwhWBIN3QyvA0qI6p25xQeuUrWE,21824
|
863
991
|
sglang/test/test_block_fp8_deep_gemm_blackwell.py,sha256=Hnhq4kkyINHb4ONedkp5Kf7Xx0MjOoOBYL8J7ETvOVA,8180
|
864
992
|
sglang/test/test_block_fp8_ep.py,sha256=n4X6ZKwuUUbV5Ofjg64ptlaFGI1LbRXDfFiJW1ELHgY,10546
|
865
993
|
sglang/test/test_custom_ops.py,sha256=PenQ8zM1wj5xwiVEPVzD37pO-x90aOfFMpCRZenaKsY,5709
|
866
|
-
sglang/test/test_cutlass_moe.py,sha256=
|
994
|
+
sglang/test/test_cutlass_moe.py,sha256=WHCruIMzxzRzxQEeFEUtk0e0IefjiO0Kjfgnx42r0UI,10069
|
867
995
|
sglang/test/test_cutlass_w4a8_moe.py,sha256=Ku9VCXEMJ3BwXtvb1A3FzY-zek-S-A4thWyX6m-1v-A,9219
|
868
996
|
sglang/test/test_deepep_utils.py,sha256=749ysTBGNzh6rYUCJhhZBtZpeD15eWTeNHYCytcvZtc,7448
|
997
|
+
sglang/test/test_deterministic.py,sha256=3TW6QogzSrvARizRRwELesKy-_NM3NfQkLEGoAvpadI,10120
|
998
|
+
sglang/test/test_disaggregation_utils.py,sha256=9LB6vCKr55iHhFN5saOwlKealG4ME20O3LTBbCCPbGI,2573
|
869
999
|
sglang/test/test_dynamic_grad_mode.py,sha256=L76yUCuk_ymNpXD2CmO8r2GiGjIvD_gtTsuFDs2NolI,1638
|
870
|
-
sglang/test/test_fp4_moe.py,sha256=
|
1000
|
+
sglang/test/test_fp4_moe.py,sha256=wvHbGMH1UBr_XVWYgX57Y2bLy9h2DPQH10_Hmf2q6iw,22436
|
871
1001
|
sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg,3764
|
872
1002
|
sglang/test/test_marlin_moe.py,sha256=TOuYlVMx2Lv_IN3AMWsg5aGaxQF472E1B9ev_vvfU2Q,7971
|
873
1003
|
sglang/test/test_marlin_utils.py,sha256=kC7IzN364mJT9xw0nKUP1V1OK1hkw6fQTBvD4nRaOz4,5151
|
874
|
-
sglang/test/test_programs.py,sha256=
|
875
|
-
sglang/test/test_utils.py,sha256=
|
1004
|
+
sglang/test/test_programs.py,sha256=R-pzgNB-h_jjdJiW50hn83-nY7z5q5XcqPwugPnVF5Y,18916
|
1005
|
+
sglang/test/test_utils.py,sha256=ED7Kktyg5tgaOv8UAFCg29XKMz0MLd7Zgl8ctQFwHOQ,54542
|
876
1006
|
sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
877
1007
|
sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
|
878
1008
|
sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
|
879
1009
|
sglang/test/attention/test_prefix_chunk_info.py,sha256=hpoDe2wfSa6RlUbfyri_c0iyBTb35UXGL9I2Xh6jamM,7772
|
880
|
-
sglang/test/attention/test_trtllm_mla_backend.py,sha256=
|
881
|
-
sglang-0.5.
|
882
|
-
sglang-0.5.
|
883
|
-
sglang-0.5.
|
884
|
-
sglang-0.5.
|
885
|
-
sglang-0.5.
|
1010
|
+
sglang/test/attention/test_trtllm_mla_backend.py,sha256=wp2_LSTwQ_vIdNPUCbEz7PA4doDU8sTWs6s6XVI0YnI,48874
|
1011
|
+
sglang-0.5.3.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
|
1012
|
+
sglang-0.5.3.dist-info/METADATA,sha256=yZukiTO_XQbUyUS98XdygJxC7qIihy6tKkB3LCbquTY,24606
|
1013
|
+
sglang-0.5.3.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
|
1014
|
+
sglang-0.5.3.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
|
1015
|
+
sglang-0.5.3.dist-info/RECORD,,
|