LICENSE
README.md
pyproject.toml
pyproject_cpu.toml
pyproject_other.toml
pyproject_xpu.toml
sglang/README.md
sglang/__init__.py
sglang/_version.py
sglang/bench_offline_throughput.py
sglang/bench_one_batch.py
sglang/bench_one_batch_server.py
sglang/bench_serving.py
sglang/check_env.py
sglang/compile_deep_gemm.py
sglang/global_config.py
sglang/launch_server.py
sglang/profiler.py
sglang/utils.py
sglang/version.py
sglang.egg-info/PKG-INFO
sglang.egg-info/SOURCES.txt
sglang.egg-info/dependency_links.txt
sglang.egg-info/entry_points.txt
sglang.egg-info/requires.txt
sglang.egg-info/top_level.txt
sglang/cli/__init__.py
sglang/cli/generate.py
sglang/cli/main.py
sglang/cli/serve.py
sglang/cli/utils.py
sglang/eval/llama3_eval.py
sglang/eval/loogle_eval.py
sglang/jit_kernel/.clang-format
sglang/jit_kernel/__main__.py
sglang/jit_kernel/add_constant.py
sglang/jit_kernel/cuda_wait_value.py
sglang/jit_kernel/hicache.py
sglang/jit_kernel/norm.py
sglang/jit_kernel/per_tensor_quant_fp8.py
sglang/jit_kernel/utils.py
sglang/jit_kernel/benchmark/bench_per_tensor_quant_fp8.py
sglang/jit_kernel/benchmark/bench_qknorm.py
sglang/jit_kernel/csrc/add_constant.cuh
sglang/jit_kernel/csrc/cuda_wait_value.cuh
sglang/jit_kernel/csrc/hicache.cuh
sglang/jit_kernel/csrc/norm.cuh
sglang/jit_kernel/csrc/test_utils.h
sglang/jit_kernel/csrc/gemm/per_tensor_quant_fp8.cuh
sglang/jit_kernel/include/sgl_kernel/fp8_utils.cuh
sglang/jit_kernel/include/sgl_kernel/runtime.cuh
sglang/jit_kernel/include/sgl_kernel/tensor.h
sglang/jit_kernel/include/sgl_kernel/utils.cuh
sglang/jit_kernel/include/sgl_kernel/utils.h
sglang/jit_kernel/include/sgl_kernel/warp.cuh
sglang/jit_kernel/tests/test_add_constant.py
sglang/jit_kernel/tests/test_per_tensor_quant_fp8.py
sglang/jit_kernel/tests/test_qknorm.py
sglang/lang/api.py
sglang/lang/chat_template.py
sglang/lang/choices.py
sglang/lang/interpreter.py
sglang/lang/ir.py
sglang/lang/tracer.py
sglang/lang/backend/anthropic.py
sglang/lang/backend/base_backend.py
sglang/lang/backend/litellm.py
sglang/lang/backend/openai.py
sglang/lang/backend/runtime_endpoint.py
sglang/lang/backend/vertexai.py
sglang/multimodal_gen/README.md
sglang/multimodal_gen/__init__.py
sglang/multimodal_gen/envs.py
sglang/multimodal_gen/registry.py
sglang/multimodal_gen/utils.py
sglang/multimodal_gen/apps/webui/README.md
sglang/multimodal_gen/apps/webui/__init__.py
sglang/multimodal_gen/apps/webui/main.py
sglang/multimodal_gen/benchmarks/bench_serving.py
sglang/multimodal_gen/benchmarks/compare_perf.py
sglang/multimodal_gen/configs/__init__.py
sglang/multimodal_gen/configs/utils.py
sglang/multimodal_gen/configs/backend/vmoba/wan_1.3B_77_448_832.json
sglang/multimodal_gen/configs/backend/vmoba/wan_1.3B_77_480_832.json
sglang/multimodal_gen/configs/models/__init__.py
sglang/multimodal_gen/configs/models/base.py
sglang/multimodal_gen/configs/models/dits/__init__.py
sglang/multimodal_gen/configs/models/dits/base.py
sglang/multimodal_gen/configs/models/dits/flux.py
sglang/multimodal_gen/configs/models/dits/hunyuanvideo.py
sglang/multimodal_gen/configs/models/dits/qwenimage.py
sglang/multimodal_gen/configs/models/dits/wanvideo.py
sglang/multimodal_gen/configs/models/dits/zimage.py
sglang/multimodal_gen/configs/models/encoders/__init__.py
sglang/multimodal_gen/configs/models/encoders/base.py
sglang/multimodal_gen/configs/models/encoders/clip.py
sglang/multimodal_gen/configs/models/encoders/llama.py
sglang/multimodal_gen/configs/models/encoders/qwen_image.py
sglang/multimodal_gen/configs/models/encoders/t5.py
sglang/multimodal_gen/configs/models/vaes/__init__.py
sglang/multimodal_gen/configs/models/vaes/base.py
sglang/multimodal_gen/configs/models/vaes/flux.py
sglang/multimodal_gen/configs/models/vaes/hunyuanvae.py
sglang/multimodal_gen/configs/models/vaes/qwenimage.py
sglang/multimodal_gen/configs/models/vaes/wanvae.py
sglang/multimodal_gen/configs/pipeline_configs/__init__.py
sglang/multimodal_gen/configs/pipeline_configs/base.py
sglang/multimodal_gen/configs/pipeline_configs/flux.py
sglang/multimodal_gen/configs/pipeline_configs/flux_finetuned.py
sglang/multimodal_gen/configs/pipeline_configs/hunyuan.py
sglang/multimodal_gen/configs/pipeline_configs/qwen_image.py
sglang/multimodal_gen/configs/pipeline_configs/wan.py
sglang/multimodal_gen/configs/pipeline_configs/zimage.py
sglang/multimodal_gen/configs/sample/__init__.py
sglang/multimodal_gen/configs/sample/flux.py
sglang/multimodal_gen/configs/sample/hunyuan.py
sglang/multimodal_gen/configs/sample/qwenimage.py
sglang/multimodal_gen/configs/sample/sampling_params.py
sglang/multimodal_gen/configs/sample/teacache.py
sglang/multimodal_gen/configs/sample/wan.py
sglang/multimodal_gen/configs/sample/zimage.py
sglang/multimodal_gen/csrc/attn/vmoba_attn/README.md
sglang/multimodal_gen/csrc/attn/vmoba_attn/setup.py
sglang/multimodal_gen/csrc/attn/vmoba_attn/tests/test_vmoba_attn.py
sglang/multimodal_gen/csrc/attn/vmoba_attn/vmoba/__init__.py
sglang/multimodal_gen/csrc/attn/vmoba_attn/vmoba/vmoba.py
sglang/multimodal_gen/docs/attention_backends.md
sglang/multimodal_gen/docs/cache_dit.md
sglang/multimodal_gen/docs/cli.md
sglang/multimodal_gen/docs/contributing.md
sglang/multimodal_gen/docs/environment_variables.md
sglang/multimodal_gen/docs/install.md
sglang/multimodal_gen/docs/install_rocm.md
sglang/multimodal_gen/docs/openai_api.md
sglang/multimodal_gen/docs/profiling.md
sglang/multimodal_gen/docs/support_matrix.md
sglang/multimodal_gen/docs/support_new_models.md
sglang/multimodal_gen/runtime/launch_server.py
sglang/multimodal_gen/runtime/scheduler_client.py
sglang/multimodal_gen/runtime/server_args.py
sglang/multimodal_gen/runtime/distributed/__init__.py
sglang/multimodal_gen/runtime/distributed/communication_op.py
sglang/multimodal_gen/runtime/distributed/group_coordinator.py
sglang/multimodal_gen/runtime/distributed/parallel_state.py
sglang/multimodal_gen/runtime/distributed/utils.py
sglang/multimodal_gen/runtime/distributed/yunchang.py
sglang/multimodal_gen/runtime/distributed/device_communicators/__init__.py
sglang/multimodal_gen/runtime/distributed/device_communicators/base_device_communicator.py
sglang/multimodal_gen/runtime/distributed/device_communicators/cpu_communicator.py
sglang/multimodal_gen/runtime/distributed/device_communicators/cuda_communicator.py
sglang/multimodal_gen/runtime/distributed/device_communicators/pynccl.py
sglang/multimodal_gen/runtime/distributed/device_communicators/pynccl_wrapper.py
sglang/multimodal_gen/runtime/entrypoints/__init__.py
sglang/multimodal_gen/runtime/entrypoints/diffusion_generator.py
sglang/multimodal_gen/runtime/entrypoints/http_server.py
sglang/multimodal_gen/runtime/entrypoints/utils.py
sglang/multimodal_gen/runtime/entrypoints/cli/__init__.py
sglang/multimodal_gen/runtime/entrypoints/cli/cli_types.py
sglang/multimodal_gen/runtime/entrypoints/cli/generate.py
sglang/multimodal_gen/runtime/entrypoints/cli/main.py
sglang/multimodal_gen/runtime/entrypoints/cli/serve.py
sglang/multimodal_gen/runtime/entrypoints/cli/utils.py
sglang/multimodal_gen/runtime/entrypoints/openai/common_api.py
sglang/multimodal_gen/runtime/entrypoints/openai/image_api.py
sglang/multimodal_gen/runtime/entrypoints/openai/protocol.py
sglang/multimodal_gen/runtime/entrypoints/openai/stores.py
sglang/multimodal_gen/runtime/entrypoints/openai/utils.py
sglang/multimodal_gen/runtime/entrypoints/openai/video_api.py
sglang/multimodal_gen/runtime/layers/__init__.py
sglang/multimodal_gen/runtime/layers/activation.py
sglang/multimodal_gen/runtime/layers/custom_op.py
sglang/multimodal_gen/runtime/layers/layernorm.py
sglang/multimodal_gen/runtime/layers/linear.py
sglang/multimodal_gen/runtime/layers/mlp.py
sglang/multimodal_gen/runtime/layers/rotary_embedding.py
sglang/multimodal_gen/runtime/layers/triton_ops.py
sglang/multimodal_gen/runtime/layers/usp.py
sglang/multimodal_gen/runtime/layers/utils.py
sglang/multimodal_gen/runtime/layers/visual_embedding.py
sglang/multimodal_gen/runtime/layers/vocab_parallel_embedding.py
sglang/multimodal_gen/runtime/layers/attention/STA_configuration.py
sglang/multimodal_gen/runtime/layers/attention/__init__.py
sglang/multimodal_gen/runtime/layers/attention/layer.py
sglang/multimodal_gen/runtime/layers/attention/selector.py
sglang/multimodal_gen/runtime/layers/attention/turbo_layer.py
sglang/multimodal_gen/runtime/layers/attention/backends/__init__.py
sglang/multimodal_gen/runtime/layers/attention/backends/aiter.py
sglang/multimodal_gen/runtime/layers/attention/backends/attention_backend.py
sglang/multimodal_gen/runtime/layers/attention/backends/flash_attn.py
sglang/multimodal_gen/runtime/layers/attention/backends/flash_attn_2.py
sglang/multimodal_gen/runtime/layers/attention/backends/sage_attn.py
sglang/multimodal_gen/runtime/layers/attention/backends/sage_attn3.py
sglang/multimodal_gen/runtime/layers/attention/backends/sdpa.py
sglang/multimodal_gen/runtime/layers/attention/backends/sliding_tile_attn.py
sglang/multimodal_gen/runtime/layers/attention/backends/video_sparse_attn.py
sglang/multimodal_gen/runtime/layers/attention/backends/vmoba.py
sglang/multimodal_gen/runtime/layers/lora/linear.py
sglang/multimodal_gen/runtime/layers/quantization/__init__.py
sglang/multimodal_gen/runtime/layers/quantization/base_config.py
sglang/multimodal_gen/runtime/loader/component_loader.py
sglang/multimodal_gen/runtime/loader/fsdp_load.py
sglang/multimodal_gen/runtime/loader/utils.py
sglang/multimodal_gen/runtime/loader/weight_utils.py
sglang/multimodal_gen/runtime/managers/forward_context.py
sglang/multimodal_gen/runtime/managers/gpu_worker.py
sglang/multimodal_gen/runtime/managers/scheduler.py
sglang/multimodal_gen/runtime/models/__init__.py
sglang/multimodal_gen/runtime/models/parameter.py
sglang/multimodal_gen/runtime/models/registry.py
sglang/multimodal_gen/runtime/models/utils.py
sglang/multimodal_gen/runtime/models/vision_utils.py
sglang/multimodal_gen/runtime/models/dits/base.py
sglang/multimodal_gen/runtime/models/dits/causal_wanvideo.py
sglang/multimodal_gen/runtime/models/dits/flux.py
sglang/multimodal_gen/runtime/models/dits/flux_2.py
sglang/multimodal_gen/runtime/models/dits/hunyuanvideo.py
sglang/multimodal_gen/runtime/models/dits/qwen_image.py
sglang/multimodal_gen/runtime/models/dits/wanvideo.py
sglang/multimodal_gen/runtime/models/dits/zimage.py
sglang/multimodal_gen/runtime/models/encoders/base.py
sglang/multimodal_gen/runtime/models/encoders/bert.py
sglang/multimodal_gen/runtime/models/encoders/clip.py
sglang/multimodal_gen/runtime/models/encoders/llama.py
sglang/multimodal_gen/runtime/models/encoders/mistral_3.py
sglang/multimodal_gen/runtime/models/encoders/qwen2_5vl.py
sglang/multimodal_gen/runtime/models/encoders/t5.py
sglang/multimodal_gen/runtime/models/encoders/vision.py
sglang/multimodal_gen/runtime/models/model_stages/qwen_image_layered.py
sglang/multimodal_gen/runtime/models/schedulers/base.py
sglang/multimodal_gen/runtime/models/schedulers/scheduling_flow_match_euler_discrete.py
sglang/multimodal_gen/runtime/models/schedulers/scheduling_flow_unipc_multistep.py
sglang/multimodal_gen/runtime/models/schedulers/scheduling_self_forcing_flow_match.py
sglang/multimodal_gen/runtime/models/schedulers/scheduling_unipc_multistep.py
sglang/multimodal_gen/runtime/models/vaes/autoencoder.py
sglang/multimodal_gen/runtime/models/vaes/autoencoder_kl_flux2.py
sglang/multimodal_gen/runtime/models/vaes/autoencoder_kl_qwenimage.py
sglang/multimodal_gen/runtime/models/vaes/common.py
sglang/multimodal_gen/runtime/models/vaes/hunyuanvae.py
sglang/multimodal_gen/runtime/models/vaes/wanvae.py
sglang/multimodal_gen/runtime/pipelines/__init__.py
sglang/multimodal_gen/runtime/pipelines/flux.py
sglang/multimodal_gen/runtime/pipelines/flux_2.py
sglang/multimodal_gen/runtime/pipelines/hunyuan_pipeline.py
sglang/multimodal_gen/runtime/pipelines/qwen_image.py
sglang/multimodal_gen/runtime/pipelines/wan_causal_dmd_pipeline.py
sglang/multimodal_gen/runtime/pipelines/wan_dmd_pipeline.py
sglang/multimodal_gen/runtime/pipelines/wan_i2v_dmd_pipeline.py
sglang/multimodal_gen/runtime/pipelines/wan_i2v_pipeline.py
sglang/multimodal_gen/runtime/pipelines/wan_pipeline.py
sglang/multimodal_gen/runtime/pipelines/zimage_pipeline.py
sglang/multimodal_gen/runtime/pipelines_core/__init__.py
sglang/multimodal_gen/runtime/pipelines_core/composed_pipeline_base.py
sglang/multimodal_gen/runtime/pipelines_core/lora_format_adapter.py
sglang/multimodal_gen/runtime/pipelines_core/lora_pipeline.py
sglang/multimodal_gen/runtime/pipelines_core/schedule_batch.py
sglang/multimodal_gen/runtime/pipelines_core/executors/parallel_executor.py
sglang/multimodal_gen/runtime/pipelines_core/executors/pipeline_executor.py
sglang/multimodal_gen/runtime/pipelines_core/executors/sync_executor.py
sglang/multimodal_gen/runtime/pipelines_core/stages/__init__.py
sglang/multimodal_gen/runtime/pipelines_core/stages/base.py
sglang/multimodal_gen/runtime/pipelines_core/stages/causal_denoising.py
sglang/multimodal_gen/runtime/pipelines_core/stages/conditioning.py
sglang/multimodal_gen/runtime/pipelines_core/stages/decoding.py
sglang/multimodal_gen/runtime/pipelines_core/stages/denoising.py
sglang/multimodal_gen/runtime/pipelines_core/stages/denoising_dmd.py
sglang/multimodal_gen/runtime/pipelines_core/stages/encoding.py
sglang/multimodal_gen/runtime/pipelines_core/stages/image_encoding.py
sglang/multimodal_gen/runtime/pipelines_core/stages/input_validation.py
sglang/multimodal_gen/runtime/pipelines_core/stages/latent_preparation.py
sglang/multimodal_gen/runtime/pipelines_core/stages/text_encoding.py
sglang/multimodal_gen/runtime/pipelines_core/stages/timestep_preparation.py
sglang/multimodal_gen/runtime/pipelines_core/stages/validators.py
sglang/multimodal_gen/runtime/platforms/__init__.py
sglang/multimodal_gen/runtime/platforms/cpu.py
sglang/multimodal_gen/runtime/platforms/cuda.py
sglang/multimodal_gen/runtime/platforms/interface.py
sglang/multimodal_gen/runtime/platforms/mps.py
sglang/multimodal_gen/runtime/platforms/rocm.py
sglang/multimodal_gen/runtime/utils/cache_dit_integration.py
sglang/multimodal_gen/runtime/utils/common.py
sglang/multimodal_gen/runtime/utils/distributed.py
sglang/multimodal_gen/runtime/utils/hf_diffusers_utils.py
sglang/multimodal_gen/runtime/utils/layerwise_offload.py
sglang/multimodal_gen/runtime/utils/logging_utils.py
sglang/multimodal_gen/runtime/utils/perf_logger.py
sglang/multimodal_gen/runtime/utils/profiler.py
sglang/multimodal_gen/test/__init__.py
sglang/multimodal_gen/test/run_suite.py
sglang/multimodal_gen/test/slack_utils.py
sglang/multimodal_gen/test/test_offline_api.py
sglang/multimodal_gen/test/test_utils.py
sglang/multimodal_gen/test/cli/test_generate_common.py
sglang/multimodal_gen/test/cli/test_generate_t2i_perf.py
sglang/multimodal_gen/test/cli/test_generate_t2v_perf.py
sglang/multimodal_gen/test/cli/test_generate_ti2v_perf.py
sglang/multimodal_gen/test/cli/test_serve.py
sglang/multimodal_gen/test/server/conftest.py
sglang/multimodal_gen/test/server/perf_baselines.json
sglang/multimodal_gen/test/server/test_lora_format_adapter.py
sglang/multimodal_gen/test/server/test_server_2_gpu_a.py
sglang/multimodal_gen/test/server/test_server_2_gpu_b.py
sglang/multimodal_gen/test/server/test_server_a.py
sglang/multimodal_gen/test/server/test_server_b.py
sglang/multimodal_gen/test/server/test_server_common.py
sglang/multimodal_gen/test/server/test_server_utils.py
sglang/multimodal_gen/test/server/testcase_configs.py
sglang/multimodal_gen/test/test_files/launch_flux.json
sglang/multimodal_gen/test/test_files/launch_wan.json
sglang/multimodal_gen/third_party/__init__.py
sglang/multimodal_gen/third_party/pynvml.py
sglang/srt/constants.py
sglang/srt/custom_op.py
sglang/srt/environ.py
sglang/srt/server_args.py
sglang/srt/server_args_config_parser.py
sglang/srt/batch_invariant_ops/__init__.py
sglang/srt/batch_invariant_ops/batch_invariant_ops.py
sglang/srt/batch_overlap/operations.py
sglang/srt/batch_overlap/operations_strategy.py
sglang/srt/batch_overlap/single_batch_overlap.py
sglang/srt/batch_overlap/two_batch_overlap.py
sglang/srt/checkpoint_engine/__init__.py
sglang/srt/checkpoint_engine/checkpoint_engine_worker.py
sglang/srt/checkpoint_engine/update.py
sglang/srt/compilation/backend.py
sglang/srt/compilation/compilation_config.py
sglang/srt/compilation/compilation_counter.py
sglang/srt/compilation/compile.py
sglang/srt/compilation/compiler_interface.py
sglang/srt/compilation/cuda_piecewise_backend.py
sglang/srt/compilation/fix_functionalization.py
sglang/srt/compilation/fx_utils.py
sglang/srt/compilation/inductor_pass.py
sglang/srt/compilation/npu_piecewise_backend.py
sglang/srt/compilation/pass_manager.py
sglang/srt/compilation/piecewise_context_manager.py
sglang/srt/compilation/weak_ref_tensor.py
sglang/srt/configs/__init__.py
sglang/srt/configs/chatglm.py
sglang/srt/configs/dbrx.py
sglang/srt/configs/deepseek_ocr.py
sglang/srt/configs/deepseekvl2.py
sglang/srt/configs/device_config.py
sglang/srt/configs/dots_ocr.py
sglang/srt/configs/dots_vlm.py
sglang/srt/configs/exaone.py
sglang/srt/configs/falcon_h1.py
sglang/srt/configs/internvl.py
sglang/srt/configs/janus_pro.py
sglang/srt/configs/jet_nemotron.py
sglang/srt/configs/jet_vlm.py
sglang/srt/configs/kimi_linear.py
sglang/srt/configs/kimi_vl.py
sglang/srt/configs/kimi_vl_moonvit.py
sglang/srt/configs/load_config.py
sglang/srt/configs/longcat_flash.py
sglang/srt/configs/mamba_utils.py
sglang/srt/configs/model_config.py
sglang/srt/configs/modelopt_config.py
sglang/srt/configs/nano_nemotron_vl.py
sglang/srt/configs/nemotron_h.py
sglang/srt/configs/olmo3.py
sglang/srt/configs/points_v15_chat.py
sglang/srt/configs/qwen3_next.py
sglang/srt/configs/qwen3_omni.py
sglang/srt/configs/qwen3_vl.py
sglang/srt/configs/radio.py
sglang/srt/configs/step3_vl.py
sglang/srt/configs/update_config.py
sglang/srt/configs/utils.py
sglang/srt/connector/__init__.py
sglang/srt/connector/base_connector.py
sglang/srt/connector/redis.py
sglang/srt/connector/remote_instance.py
sglang/srt/connector/s3.py
sglang/srt/connector/utils.py
sglang/srt/connector/serde/__init__.py
sglang/srt/connector/serde/safe_serde.py
sglang/srt/connector/serde/serde.py
sglang/srt/constrained/base_grammar_backend.py
sglang/srt/constrained/llguidance_backend.py
sglang/srt/constrained/outlines_backend.py
sglang/srt/constrained/outlines_jump_forward.py
sglang/srt/constrained/reasoner_grammar_backend.py
sglang/srt/constrained/utils.py
sglang/srt/constrained/xgrammar_backend.py
sglang/srt/constrained/triton_ops/bitmask_ops.py
sglang/srt/debug_utils/__init__.py
sglang/srt/debug_utils/dump_comparator.py
sglang/srt/debug_utils/dump_loader.py
sglang/srt/debug_utils/dumper.py
sglang/srt/debug_utils/log_parser.py
sglang/srt/debug_utils/model_truncator.py
sglang/srt/debug_utils/tensor_dump_forward_hook.py
sglang/srt/debug_utils/text_comparator.py
sglang/srt/disaggregation/decode.py
sglang/srt/disaggregation/decode_kvcache_offload_manager.py
sglang/srt/disaggregation/decode_schedule_batch_mixin.py
sglang/srt/disaggregation/encode_receiver.py
sglang/srt/disaggregation/encode_server.py
sglang/srt/disaggregation/kv_events.py
sglang/srt/disaggregation/prefill.py
sglang/srt/disaggregation/utils.py
sglang/srt/disaggregation/ascend/__init__.py
sglang/srt/disaggregation/ascend/conn.py
sglang/srt/disaggregation/ascend/transfer_engine.py
sglang/srt/disaggregation/base/__init__.py
sglang/srt/disaggregation/base/conn.py
sglang/srt/disaggregation/common/__init__.py
sglang/srt/disaggregation/common/conn.py
sglang/srt/disaggregation/common/utils.py
sglang/srt/disaggregation/fake/__init__.py
sglang/srt/disaggregation/fake/conn.py
sglang/srt/disaggregation/mooncake/__init__.py
sglang/srt/disaggregation/mooncake/conn.py
sglang/srt/disaggregation/mooncake/transfer_engine.py
sglang/srt/disaggregation/mooncake/utils.py
sglang/srt/disaggregation/nixl/__init__.py
sglang/srt/disaggregation/nixl/conn.py
sglang/srt/distributed/__init__.py
sglang/srt/distributed/communication_op.py
sglang/srt/distributed/naive_distributed.py
sglang/srt/distributed/parallel_state.py
sglang/srt/distributed/utils.py
sglang/srt/distributed/device_communicators/all_reduce_utils.py
sglang/srt/distributed/device_communicators/cuda_wrapper.py
sglang/srt/distributed/device_communicators/custom_all_reduce.py
sglang/srt/distributed/device_communicators/custom_all_reduce_ops.py
sglang/srt/distributed/device_communicators/custom_all_reduce_utils.py
sglang/srt/distributed/device_communicators/hpu_communicator.py
sglang/srt/distributed/device_communicators/npu_communicator.py
sglang/srt/distributed/device_communicators/pymscclpp.py
sglang/srt/distributed/device_communicators/pynccl.py
sglang/srt/distributed/device_communicators/pynccl_allocator.py
sglang/srt/distributed/device_communicators/pynccl_wrapper.py
sglang/srt/distributed/device_communicators/quick_all_reduce.py
sglang/srt/distributed/device_communicators/shm_broadcast.py
sglang/srt/distributed/device_communicators/torch_symm_mem.py
sglang/srt/distributed/device_communicators/xpu_communicator.py
sglang/srt/dllm/config.py
sglang/srt/dllm/algorithm/__init__.py
sglang/srt/dllm/algorithm/base.py
sglang/srt/dllm/algorithm/low_confidence.py
sglang/srt/elastic_ep/elastic_ep.py
sglang/srt/entrypoints/EngineBase.py
sglang/srt/entrypoints/context.py
sglang/srt/entrypoints/engine.py
sglang/srt/entrypoints/grpc_server.py
sglang/srt/entrypoints/harmony_utils.py
sglang/srt/entrypoints/http_server.py
sglang/srt/entrypoints/http_server_engine.py
sglang/srt/entrypoints/tool.py
sglang/srt/entrypoints/warmup.py
sglang/srt/entrypoints/ollama/README.md
sglang/srt/entrypoints/ollama/__init__.py
sglang/srt/entrypoints/ollama/protocol.py
sglang/srt/entrypoints/ollama/serving.py
sglang/srt/entrypoints/ollama/smart_router.py
sglang/srt/entrypoints/openai/__init__.py
sglang/srt/entrypoints/openai/encoding_dsv32.py
sglang/srt/entrypoints/openai/protocol.py
sglang/srt/entrypoints/openai/serving_base.py
sglang/srt/entrypoints/openai/serving_chat.py
sglang/srt/entrypoints/openai/serving_classify.py
sglang/srt/entrypoints/openai/serving_completions.py
sglang/srt/entrypoints/openai/serving_embedding.py
sglang/srt/entrypoints/openai/serving_rerank.py
sglang/srt/entrypoints/openai/serving_responses.py
sglang/srt/entrypoints/openai/serving_score.py
sglang/srt/entrypoints/openai/serving_tokenize.py
sglang/srt/entrypoints/openai/tool_server.py
sglang/srt/entrypoints/openai/usage_processor.py
sglang/srt/entrypoints/openai/utils.py
sglang/srt/eplb/__init__.py
sglang/srt/eplb/eplb_manager.py
sglang/srt/eplb/expert_distribution.py
sglang/srt/eplb/expert_location.py
sglang/srt/eplb/expert_location_dispatch.py
sglang/srt/eplb/expert_location_updater.py
sglang/srt/eplb/eplb_algorithms/__init__.py
sglang/srt/eplb/eplb_algorithms/deepseek.py
sglang/srt/eplb/eplb_algorithms/deepseek_vec.py
sglang/srt/eplb/eplb_algorithms/elasticity_aware.py
sglang/srt/eplb/eplb_simulator/__init__.py
sglang/srt/eplb/eplb_simulator/reader.py
sglang/srt/function_call/base_format_detector.py
sglang/srt/function_call/core_types.py
sglang/srt/function_call/deepseekv31_detector.py
sglang/srt/function_call/deepseekv32_detector.py
sglang/srt/function_call/deepseekv3_detector.py
sglang/srt/function_call/function_call_parser.py
sglang/srt/function_call/glm47_moe_detector.py
sglang/srt/function_call/glm4_moe_detector.py
sglang/srt/function_call/gpt_oss_detector.py
sglang/srt/function_call/internlm_detector.py
sglang/srt/function_call/json_array_parser.py
sglang/srt/function_call/kimik2_detector.py
sglang/srt/function_call/llama32_detector.py
sglang/srt/function_call/mimo_detector.py
sglang/srt/function_call/minimax_m2.py
sglang/srt/function_call/mistral_detector.py
sglang/srt/function_call/pythonic_detector.py
sglang/srt/function_call/qwen25_detector.py
sglang/srt/function_call/qwen3_coder_detector.py
sglang/srt/function_call/step3_detector.py
sglang/srt/function_call/utils.py
sglang/srt/grpc/__init__.py
sglang/srt/grpc/compile_proto.py
sglang/srt/grpc/grpc_request_manager.py
sglang/srt/grpc/health_servicer.py
sglang/srt/grpc/scheduler_launcher.py
sglang/srt/grpc/sglang_scheduler.proto
sglang/srt/grpc/sglang_scheduler_pb2.py
sglang/srt/grpc/sglang_scheduler_pb2.pyi
sglang/srt/grpc/sglang_scheduler_pb2_grpc.py
sglang/srt/hardware_backend/npu/allocator_npu.py
sglang/srt/hardware_backend/npu/cmo.py
sglang/srt/hardware_backend/npu/memory_pool_npu.py
sglang/srt/hardware_backend/npu/utils.py
sglang/srt/hardware_backend/npu/attention/ascend_backend.py
sglang/srt/hardware_backend/npu/attention/mla_preprocess.py
sglang/srt/hardware_backend/npu/graph_runner/eagle_draft_extend_npu_graph_runner.py
sglang/srt/hardware_backend/npu/graph_runner/eagle_draft_npu_graph_runner.py
sglang/srt/hardware_backend/npu/graph_runner/npu_graph_runner.py
sglang/srt/hardware_backend/npu/modules/deepseek_v2_attention_mla_npu.py
sglang/srt/hardware_backend/npu/moe/topk.py
sglang/srt/hardware_backend/npu/quantization/fused_moe_method_npu.py
sglang/srt/hardware_backend/npu/quantization/linear_method_npu.py
sglang/srt/hardware_backend/npu/quantization/modelslim.py
sglang/srt/layers/activation.py
sglang/srt/layers/amx_utils.py
sglang/srt/layers/communicator.py
sglang/srt/layers/communicator_nsa_cp.py
sglang/srt/layers/dp_attention.py
sglang/srt/layers/elementwise.py
sglang/srt/layers/flashinfer_comm_fusion.py
sglang/srt/layers/layernorm.py
sglang/srt/layers/linear.py
sglang/srt/layers/logits_processor.py
sglang/srt/layers/model_parallel.py
sglang/srt/layers/modelopt_utils.py
sglang/srt/layers/multimodal.py
sglang/srt/layers/parameter.py
sglang/srt/layers/pooler.py
sglang/srt/layers/radix_attention.py
sglang/srt/layers/rocm_linear_utils.py
sglang/srt/layers/rotary_embedding.py
sglang/srt/layers/sampler.py
sglang/srt/layers/sparse_pooler.py
sglang/srt/layers/torchao_utils.py
sglang/srt/layers/vocab_parallel_embedding.py
sglang/srt/layers/attention/aiter_backend.py
sglang/srt/layers/attention/attention_registry.py
sglang/srt/layers/attention/base_attn_backend.py
sglang/srt/layers/attention/cutlass_mla_backend.py
sglang/srt/layers/attention/double_sparsity_backend.py
sglang/srt/layers/attention/dual_chunk_flashattention_backend.py
sglang/srt/layers/attention/flashattention_backend.py
sglang/srt/layers/attention/flashinfer_backend.py
sglang/srt/layers/attention/flashinfer_mla_backend.py
sglang/srt/layers/attention/flashmla_backend.py
sglang/srt/layers/attention/hybrid_attn_backend.py
sglang/srt/layers/attention/hybrid_linear_attn_backend.py
sglang/srt/layers/attention/intel_amx_backend.py
sglang/srt/layers/attention/merge_state.py
sglang/srt/layers/attention/nsa_backend.py
sglang/srt/layers/attention/tbo_backend.py
sglang/srt/layers/attention/torch_flex_backend.py
sglang/srt/layers/attention/torch_native_backend.py
sglang/srt/layers/attention/triton_backend.py
sglang/srt/layers/attention/trtllm_mha_backend.py
sglang/srt/layers/attention/trtllm_mla_backend.py
sglang/srt/layers/attention/utils.py
sglang/srt/layers/attention/vision.py
sglang/srt/layers/attention/vision_utils.py
sglang/srt/layers/attention/wave_backend.py
sglang/srt/layers/attention/xpu_backend.py
sglang/srt/layers/attention/fla/chunk.py
sglang/srt/layers/attention/fla/chunk_delta_h.py
sglang/srt/layers/attention/fla/chunk_o.py
sglang/srt/layers/attention/fla/chunk_scaled_dot_kkt.py
sglang/srt/layers/attention/fla/cumsum.py
sglang/srt/layers/attention/fla/fused_gdn_gating.py
sglang/srt/layers/attention/fla/fused_recurrent.py
sglang/srt/layers/attention/fla/fused_sigmoid_gating_recurrent.py
sglang/srt/layers/attention/fla/index.py
sglang/srt/layers/attention/fla/kda.py
sglang/srt/layers/attention/fla/l2norm.py
sglang/srt/layers/attention/fla/layernorm_gated.py
sglang/srt/layers/attention/fla/op.py
sglang/srt/layers/attention/fla/solve_tril.py
sglang/srt/layers/attention/fla/utils.py
sglang/srt/layers/attention/fla/wy_fast.py
sglang/srt/layers/attention/mamba/causal_conv1d.py
sglang/srt/layers/attention/mamba/causal_conv1d_triton.py
sglang/srt/layers/attention/mamba/mamba.py
sglang/srt/layers/attention/mamba/mamba2_metadata.py
sglang/srt/layers/attention/mamba/mixer2_rms_norm_gated.py
sglang/srt/layers/attention/mamba/ops/__init__.py
sglang/srt/layers/attention/mamba/ops/layernorm_gated.py
sglang/srt/layers/attention/mamba/ops/mamba_ssm.py
sglang/srt/layers/attention/mamba/ops/ssd_bmm.py
sglang/srt/layers/attention/mamba/ops/ssd_chunk_scan.py
sglang/srt/layers/attention/mamba/ops/ssd_chunk_state.py
sglang/srt/layers/attention/mamba/ops/ssd_combined.py
sglang/srt/layers/attention/mamba/ops/ssd_state_passing.py
sglang/srt/layers/attention/nsa/dequant_k_cache.py
sglang/srt/layers/attention/nsa/index_buf_accessor.py
sglang/srt/layers/attention/nsa/nsa_backend_mtp_precompute.py
sglang/srt/layers/attention/nsa/nsa_indexer.py
sglang/srt/layers/attention/nsa/quant_k_cache.py
sglang/srt/layers/attention/nsa/tilelang_kernel.py
sglang/srt/layers/attention/nsa/transform_index.py
sglang/srt/layers/attention/nsa/triton_kernel.py
sglang/srt/layers/attention/nsa/utils.py
sglang/srt/layers/attention/triton_ops/decode_attention.py
sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py
sglang/srt/layers/attention/triton_ops/extend_attention.py
sglang/srt/layers/attention/triton_ops/merge_state.py
sglang/srt/layers/attention/triton_ops/prefill_attention.py
sglang/srt/layers/attention/triton_ops/rocm_mla_decode_rope.py
sglang/srt/layers/attention/triton_ops/trtllm_fp8_kv_kernel.py
sglang/srt/layers/attention/wave_ops/decode_attention.py
sglang/srt/layers/attention/wave_ops/extend_attention.py
sglang/srt/layers/attention/wave_ops/prefill_attention.py
sglang/srt/layers/deep_gemm_wrapper/__init__.py
sglang/srt/layers/deep_gemm_wrapper/compile_utils.py
sglang/srt/layers/deep_gemm_wrapper/configurer.py
sglang/srt/layers/deep_gemm_wrapper/entrypoint.py
sglang/srt/layers/moe/__init__.py
sglang/srt/layers/moe/cutlass_moe.py
sglang/srt/layers/moe/cutlass_moe_params.py
sglang/srt/layers/moe/cutlass_w4a8_moe.py
sglang/srt/layers/moe/flashinfer_cutedsl_moe.py
sglang/srt/layers/moe/fused_moe_native.py
sglang/srt/layers/moe/kt_ep_wrapper.py
sglang/srt/layers/moe/rocm_moe_utils.py
sglang/srt/layers/moe/routed_experts_capturer.py
sglang/srt/layers/moe/router.py
sglang/srt/layers/moe/topk.py
sglang/srt/layers/moe/utils.py
sglang/srt/layers/moe/ep_moe/__init__.py
sglang/srt/layers/moe/ep_moe/kernels.py
sglang/srt/layers/moe/ep_moe/layer.py
sglang/srt/layers/moe/fused_moe_triton/__init__.py
sglang/srt/layers/moe/fused_moe_triton/fused_marlin_moe.py
sglang/srt/layers/moe/fused_moe_triton/fused_moe.py
sglang/srt/layers/moe/fused_moe_triton/fused_moe_triton_config.py
sglang/srt/layers/moe/fused_moe_triton/fused_moe_triton_kernels.py
sglang/srt/layers/moe/fused_moe_triton/layer.py
sglang/srt/layers/moe/fused_moe_triton/moe_align_block_size.py
sglang/srt/layers/moe/fused_moe_triton/triton_kernels_moe.py
sglang/srt/layers/moe/fused_moe_triton/configs/README.md
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=3072,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=3072,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=3072,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=144,N=512,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=1024,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=1344,device_name=NVIDIA_A100-SXM4-40GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=1344,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=1344,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=2688,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=2688,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=3072,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=3072,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=3200,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=3584,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=6400,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=16,N=800,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=160,N=192,device_name=NVIDIA_A800-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=20,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=24,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=128,device_name=NVIDIA_H20,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=128,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=256,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=256,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=256,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=256,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=64,device_name=NVIDIA_A800-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=64,device_name=NVIDIA_L20,dtype=int8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=256,N=64,device_name=NVIDIA_L40S,dtype=int8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=1024,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=1280,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=1280,device_name=NVIDIA_A800-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=1280,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=1280,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=1280,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=1280,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=2560,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=2560,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=2560,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=320,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=320,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=320,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=320,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=640,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=640,device_name=NVIDIA_A800-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=640,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=640,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=640,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=640,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=64,N=640,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=14336,device_name=AMD_Instinct_MI300X.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=14336,device_name=AMD_Instinct_MI325X.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=14336,device_name=AMD_Radeon_Graphics.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=14336,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=14336,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=14336,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=1792,device_name=AMD_Instinct_MI300X.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=1792,device_name=AMD_Instinct_MI325X.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=1792,device_name=AMD_Radeon_Graphics.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=1792,device_name=NVIDIA_A100-SXM4-40GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=1792,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=1792,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=1792,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=2048,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=2048,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=2048,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=2048,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=AMD_Instinct_MI300X.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=AMD_Instinct_MI325X.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=AMD_Radeon_Graphics.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=NVIDIA_A100-SXM4-40GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=3584,device_name=NVIDIA_L40S.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=4096,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=4096,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=4096,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=4096,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=4096,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=4096,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=4096,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=4096,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=7168,device_name=AMD_Instinct_MI300X.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=7168,device_name=AMD_Instinct_MI325X.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=7168,device_name=AMD_Radeon_Graphics.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=7168,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=8192,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=8192,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=8192,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=8192,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_1_0/E=8,N=8192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=192,device_name=NVIDIA_A800-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=192,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=192,device_name=NVIDIA_H20.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=192,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=384,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=384,device_name=NVIDIA_H20.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=384,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=512,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=768,device_name=NVIDIA_A800-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=768,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=768,device_name=NVIDIA_H20.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=768,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=768,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=128,N=96,device_name=NVIDIA_H20.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=129,N=352,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=160,N=320,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=161,N=192,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=257,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=264,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=264,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=264,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=264,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=272,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=272,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=272,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=272,N=64,device_name=NVIDIA_A800-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=288,N=64,device_name=NVIDIA_A800-SXM4-80GB.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_0/E=16,N=1024,device_name=NVIDIA_B200.json
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
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
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
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=128,N=768,device_name=NVIDIA_H20.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=320,device_name=NVIDIA_H20-3e.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=160,N=640,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=128,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_H20-3e,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=257,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=384,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=384,N=128,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=384,N=256,device_name=NVIDIA_H20-3e,dtype=fp8_w8a8,block_shape=[128, 128].json
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
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=1856,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=1856,device_name=NVIDIA_L40S.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json
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
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=128,N=928,device_name=NVIDIA_L40S.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=129,N=352,device_name=NVIDIA_B200,dtype=fp8_w8a8.json
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=129,N=704,device_name=NVIDIA_B200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=160,N=384,device_name=NVIDIA_B200,dtype=fp8_w8a8.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=161,N=192,device_name=NVIDIA_H200,dtype=fp8_w8a8,per_channel_quant=True.json
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
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=256,N=256,device_name=NVIDIA_B200.json
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=256,N=512,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=256,N=512,device_name=NVIDIA_H20.json
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=257,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=257,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128]_down.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=257,N=64,device_name=NVIDIA_A100-SXM4-80GB.json
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H20-3e.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=128,device_name=NVIDIA_H200.json
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
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=256,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=256,device_name=NVIDIA_H20-3e.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=256,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=64,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_4_0/E=512,N=64,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=128,N=1856,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=128,N=1856,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=128,N=232,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=128,N=232,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=128,N=464,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=128,N=464,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=128,N=928,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=128,N=928,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=16,N=1856,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=16,N=1856,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=161,N=192,device_name=NVIDIA_B200,dtype=fp8_w8a8,per_channel_quant=True.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=161,N=192,device_name=NVIDIA_H200,dtype=fp8_w8a8,per_channel_quant=True.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=161,N=192,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=161,N=384,device_name=NVIDIA_B200,dtype=fp8_w8a8,per_channel_quant=True.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=161,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8,per_channel_quant=True.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=20,N=1536,device_name=NVIDIA_H200,dtype=fp8_w8a8,per_channel_quant=True.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=20,N=1536,device_name=NVIDIA_H200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=32,N=1856,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=32,N=1856,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=32,N=928,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=32,N=928,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=40,N=1536,device_name=NVIDIA_H200,dtype=fp8_w8a8,per_channel_quant=True.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=64,N=1856,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=64,N=1856,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=64,N=464,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=64,N=464,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=64,N=928,device_name=NVIDIA_B200.json
sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_1/E=64,N=928,device_name=NVIDIA_H100_80GB_HBM3.json
sglang/srt/layers/moe/moe_runner/__init__.py
sglang/srt/layers/moe/moe_runner/base.py
sglang/srt/layers/moe/moe_runner/deep_gemm.py
sglang/srt/layers/moe/moe_runner/marlin.py
sglang/srt/layers/moe/moe_runner/runner.py
sglang/srt/layers/moe/moe_runner/triton.py
sglang/srt/layers/moe/moe_runner/triton_kernels.py
sglang/srt/layers/moe/token_dispatcher/__init__.py
sglang/srt/layers/moe/token_dispatcher/base.py
sglang/srt/layers/moe/token_dispatcher/deepep.py
sglang/srt/layers/moe/token_dispatcher/fuseep.py
sglang/srt/layers/moe/token_dispatcher/mooncake.py
sglang/srt/layers/moe/token_dispatcher/standard.py
sglang/srt/layers/quantization/__init__.py
sglang/srt/layers/quantization/auto_round.py
sglang/srt/layers/quantization/awq.py
sglang/srt/layers/quantization/awq_triton.py
sglang/srt/layers/quantization/base_config.py
sglang/srt/layers/quantization/blockwise_int8.py
sglang/srt/layers/quantization/fp8.py
sglang/srt/layers/quantization/fp8_kernel.py
sglang/srt/layers/quantization/fp8_utils.py
sglang/srt/layers/quantization/fpgemm_fp8.py
sglang/srt/layers/quantization/gguf.py
sglang/srt/layers/quantization/gptq.py
sglang/srt/layers/quantization/int8_kernel.py
sglang/srt/layers/quantization/int8_utils.py
sglang/srt/layers/quantization/kv_cache.py
sglang/srt/layers/quantization/kvfp4_tensor.py
sglang/srt/layers/quantization/marlin_utils.py
sglang/srt/layers/quantization/marlin_utils_fp8.py
sglang/srt/layers/quantization/modelopt_quant.py
sglang/srt/layers/quantization/moe_wna16.py
sglang/srt/layers/quantization/mxfp4.py
sglang/srt/layers/quantization/mxfp4_tensor.py
sglang/srt/layers/quantization/petit.py
sglang/srt/layers/quantization/petit_utils.py
sglang/srt/layers/quantization/qoq.py
sglang/srt/layers/quantization/rocm_mxfp4_utils.py
sglang/srt/layers/quantization/unquant.py
sglang/srt/layers/quantization/utils.py
sglang/srt/layers/quantization/w4afp8.py
sglang/srt/layers/quantization/w8a8_fp8.py
sglang/srt/layers/quantization/w8a8_int8.py
sglang/srt/layers/quantization/compressed_tensors/README.md
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py
sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py
sglang/srt/layers/quantization/compressed_tensors/utils.py
sglang/srt/layers/quantization/compressed_tensors/schemes/__init__.py
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_scheme.py
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w4a4_nvfp4.py
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a16_fp8.py
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_fp8.py
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_int8.py
sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_wNa16.py
sglang/srt/layers/quantization/configs/N=1280,K=5120,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=1536,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=1536,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=1536,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=3072,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=36864,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=36864,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4096,K=512,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4096,K=512,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4096,K=512,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4096,K=512,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4096,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4096,K=512,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4096,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4608,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4608,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4608,K=7168,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4608,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4608,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4608,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=4608,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=512,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=512,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=512,K=7168,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=512,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=512,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=512,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=5120,K=1024,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=5120,K=3200,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=6400,K=5120,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2304,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2304,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2304,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2304,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2304,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2304,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=2304,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json
sglang/srt/layers/quantization/configs/README.md
sglang/srt/layers/quantization/quark/__init__.py
sglang/srt/layers/quantization/quark/quark.py
sglang/srt/layers/quantization/quark/quark_moe.py
sglang/srt/layers/quantization/quark/utils.py
sglang/srt/layers/quantization/quark/schemes/__init__.py
sglang/srt/layers/quantization/quark/schemes/quark_scheme.py
sglang/srt/layers/quantization/quark/schemes/quark_w4a4_mxfp4.py
sglang/srt/layers/quantization/quark/schemes/quark_w8a8_fp8.py
sglang/srt/layers/utils/__init__.py
sglang/srt/layers/utils/common.py
sglang/srt/layers/utils/logprob.py
sglang/srt/lora/eviction_policy.py
sglang/srt/lora/layers.py
sglang/srt/lora/lora.py
sglang/srt/lora/lora_config.py
sglang/srt/lora/lora_manager.py
sglang/srt/lora/lora_registry.py
sglang/srt/lora/mem_pool.py
sglang/srt/lora/utils.py
sglang/srt/lora/backend/ascend_backend.py
sglang/srt/lora/backend/base_backend.py
sglang/srt/lora/backend/chunked_backend.py
sglang/srt/lora/backend/lora_registry.py
sglang/srt/lora/backend/torch_backend.py
sglang/srt/lora/backend/triton_backend.py
sglang/srt/lora/torch_ops/__init__.py
sglang/srt/lora/torch_ops/lora_ops.py
sglang/srt/lora/triton_ops/__init__.py
sglang/srt/lora/triton_ops/chunked_sgmv_expand.py
sglang/srt/lora/triton_ops/chunked_sgmv_shrink.py
sglang/srt/lora/triton_ops/embedding_lora_a.py
sglang/srt/lora/triton_ops/gate_up_lora_b.py
sglang/srt/lora/triton_ops/qkv_lora_b.py
sglang/srt/lora/triton_ops/sgemm_lora_a.py
sglang/srt/lora/triton_ops/sgemm_lora_b.py
sglang/srt/managers/async_dynamic_batch_tokenizer.py
sglang/srt/managers/async_mm_data_processor.py
sglang/srt/managers/cache_controller.py
sglang/srt/managers/configure_logging.py
sglang/srt/managers/data_parallel_controller.py
sglang/srt/managers/detokenizer_manager.py
sglang/srt/managers/disagg_service.py
sglang/srt/managers/io_struct.py
sglang/srt/managers/mm_utils.py
sglang/srt/managers/multi_tokenizer_mixin.py
sglang/srt/managers/multimodal_processor.py
sglang/srt/managers/overlap_utils.py
sglang/srt/managers/request_metrics_exporter.py
sglang/srt/managers/schedule_batch.py
sglang/srt/managers/schedule_policy.py
sglang/srt/managers/scheduler.py
sglang/srt/managers/scheduler_dp_attn_mixin.py
sglang/srt/managers/scheduler_enhancer.py
sglang/srt/managers/scheduler_input_blocker.py
sglang/srt/managers/scheduler_metrics_mixin.py
sglang/srt/managers/scheduler_output_processor_mixin.py
sglang/srt/managers/scheduler_pp_mixin.py
sglang/srt/managers/scheduler_profiler_mixin.py
sglang/srt/managers/scheduler_recv_skipper.py
sglang/srt/managers/scheduler_runtime_checker_mixin.py
sglang/srt/managers/scheduler_update_weights_mixin.py
sglang/srt/managers/session_controller.py
sglang/srt/managers/template_manager.py
sglang/srt/managers/tokenizer_communicator_mixin.py
sglang/srt/managers/tokenizer_manager.py
sglang/srt/managers/tokenizer_manager_multiitem_mixin.py
sglang/srt/managers/tp_worker.py
sglang/srt/managers/utils.py
sglang/srt/mem_cache/allocator.py
sglang/srt/mem_cache/base_prefix_cache.py
sglang/srt/mem_cache/cache_init_params.py
sglang/srt/mem_cache/chunk_cache.py
sglang/srt/mem_cache/common.py
sglang/srt/mem_cache/evict_policy.py
sglang/srt/mem_cache/flush_cache.py
sglang/srt/mem_cache/hicache_storage.py
sglang/srt/mem_cache/hiradix_cache.py
sglang/srt/mem_cache/mamba_radix_cache.py
sglang/srt/mem_cache/memory_pool.py
sglang/srt/mem_cache/memory_pool_host.py
sglang/srt/mem_cache/multimodal_cache.py
sglang/srt/mem_cache/radix_cache.py
sglang/srt/mem_cache/radix_cache_cpp.py
sglang/srt/mem_cache/swa_radix_cache.py
sglang/srt/mem_cache/utils.py
sglang/srt/mem_cache/cpp_radix_tree/.clang-format
sglang/srt/mem_cache/cpp_radix_tree/common.h
sglang/srt/mem_cache/cpp_radix_tree/radix_tree.py
sglang/srt/mem_cache/cpp_radix_tree/tree_v2.cpp
sglang/srt/mem_cache/cpp_radix_tree/tree_v2.h
sglang/srt/mem_cache/cpp_radix_tree/tree_v2_binding.cpp
sglang/srt/mem_cache/cpp_radix_tree/tree_v2_debug.cpp
sglang/srt/mem_cache/cpp_radix_tree/tree_v2_impl.h
sglang/srt/mem_cache/cpp_radix_tree/tree_v2_node.h
sglang/srt/mem_cache/sparsity/algorithms/__init__.py
sglang/srt/mem_cache/sparsity/algorithms/base_algorithm.py
sglang/srt/mem_cache/sparsity/algorithms/deepseek_nsa.py
sglang/srt/mem_cache/sparsity/algorithms/quest_algorithm.py
sglang/srt/mem_cache/storage/__init__.py
sglang/srt/mem_cache/storage/backend_factory.py
sglang/srt/mem_cache/storage/aibrix_kvcache/README.md
sglang/srt/mem_cache/storage/aibrix_kvcache/aibrix_kvcache_storage.py
sglang/srt/mem_cache/storage/aibrix_kvcache/unit_test.py
sglang/srt/mem_cache/storage/eic/README.md
sglang/srt/mem_cache/storage/eic/eic_storage.py
sglang/srt/mem_cache/storage/eic/test_unit.py
sglang/srt/mem_cache/storage/hf3fs/hf3fs_client.py
sglang/srt/mem_cache/storage/hf3fs/hf3fs_usrbio_client.py
sglang/srt/mem_cache/storage/hf3fs/hf3fs_utils.cpp
sglang/srt/mem_cache/storage/hf3fs/mini_3fs_metadata_server.py
sglang/srt/mem_cache/storage/hf3fs/storage_hf3fs.py
sglang/srt/mem_cache/storage/hf3fs/test_hf3fs_utils.py
sglang/srt/mem_cache/storage/hf3fs/docs/README.md
sglang/srt/mem_cache/storage/hf3fs/docs/deploy_sglang_3fs_multinode.md
sglang/srt/mem_cache/storage/hf3fs/docs/setup_usrbio_client.md
sglang/srt/mem_cache/storage/lmcache/README.md
sglang/srt/mem_cache/storage/lmcache/example_config.yaml
sglang/srt/mem_cache/storage/lmcache/lmc_radix_cache.py
sglang/srt/mem_cache/storage/lmcache/unit_test.py
sglang/srt/mem_cache/storage/mooncake_store/README.md
sglang/srt/mem_cache/storage/mooncake_store/mooncake_store.py
sglang/srt/mem_cache/storage/mooncake_store/test_mooncake_store.py
sglang/srt/mem_cache/storage/nixl/README.md
sglang/srt/mem_cache/storage/nixl/hicache_nixl.py
sglang/srt/mem_cache/storage/nixl/nixl_utils.py
sglang/srt/mem_cache/storage/nixl/test_hicache_nixl_storage.py
sglang/srt/metrics/collector.py
sglang/srt/metrics/func_timer.py
sglang/srt/metrics/label_transform.py
sglang/srt/metrics/startup_func_log_and_timer.py
sglang/srt/metrics/utils.py
sglang/srt/model_executor/cpu_graph_runner.py
sglang/srt/model_executor/cuda_graph_runner.py
sglang/srt/model_executor/forward_batch_info.py
sglang/srt/model_executor/hook_manager.py
sglang/srt/model_executor/input_buffers.py
sglang/srt/model_executor/mindspore_runner.py
sglang/srt/model_executor/model_runner.py
sglang/srt/model_executor/model_runner_kv_cache_mixin.py
sglang/srt/model_executor/piecewise_cuda_graph_runner.py
sglang/srt/model_loader/__init__.py
sglang/srt/model_loader/ci_weight_validation.py
sglang/srt/model_loader/loader.py
sglang/srt/model_loader/remote_instance_weight_loader_utils.py
sglang/srt/model_loader/utils.py
sglang/srt/model_loader/weight_utils.py
sglang/srt/models/apertus.py
sglang/srt/models/arcee.py
sglang/srt/models/baichuan.py
sglang/srt/models/bailing_moe.py
sglang/srt/models/bailing_moe_nextn.py
sglang/srt/models/bert.py
sglang/srt/models/chatglm.py
sglang/srt/models/clip.py
sglang/srt/models/commandr.py
sglang/srt/models/dbrx.py
sglang/srt/models/deepseek.py
sglang/srt/models/deepseek_janus_pro.py
sglang/srt/models/deepseek_nextn.py
sglang/srt/models/deepseek_ocr.py
sglang/srt/models/deepseek_v2.py
sglang/srt/models/deepseek_vl2.py
sglang/srt/models/dots_ocr.py
sglang/srt/models/dots_vlm.py
sglang/srt/models/dots_vlm_vit.py
sglang/srt/models/ernie4.py
sglang/srt/models/ernie4_eagle.py
sglang/srt/models/exaone.py
sglang/srt/models/falcon_h1.py
sglang/srt/models/gemma.py
sglang/srt/models/gemma2.py
sglang/srt/models/gemma2_reward.py
sglang/srt/models/gemma3_causal.py
sglang/srt/models/gemma3_mm.py
sglang/srt/models/gemma3n_audio.py
sglang/srt/models/gemma3n_causal.py
sglang/srt/models/gemma3n_mm.py
sglang/srt/models/glm4.py
sglang/srt/models/glm4_moe.py
sglang/srt/models/glm4_moe_nextn.py
sglang/srt/models/glm4v.py
sglang/srt/models/glm4v_moe.py
sglang/srt/models/glmasr.py
sglang/srt/models/gpt2.py
sglang/srt/models/gpt_bigcode.py
sglang/srt/models/gpt_oss.py
sglang/srt/models/granite.py
sglang/srt/models/granitemoe.py
sglang/srt/models/grok.py
sglang/srt/models/hunyuan.py
sglang/srt/models/idefics2.py
sglang/srt/models/internlm2.py
sglang/srt/models/internlm2_reward.py
sglang/srt/models/interns1.py
sglang/srt/models/internvl.py
sglang/srt/models/jet_nemotron.py
sglang/srt/models/jet_vlm.py
sglang/srt/models/kimi_linear.py
sglang/srt/models/kimi_vl.py
sglang/srt/models/kimi_vl_moonvit.py
sglang/srt/models/llada2.py
sglang/srt/models/llama.py
sglang/srt/models/llama4.py
sglang/srt/models/llama_classification.py
sglang/srt/models/llama_eagle.py
sglang/srt/models/llama_eagle3.py
sglang/srt/models/llama_embedding.py
sglang/srt/models/llama_reward.py
sglang/srt/models/llava.py
sglang/srt/models/llavavid.py
sglang/srt/models/longcat_flash.py
sglang/srt/models/longcat_flash_nextn.py
sglang/srt/models/midashenglm.py
sglang/srt/models/mimo.py
sglang/srt/models/mimo_mtp.py
sglang/srt/models/mimo_v2_flash.py
sglang/srt/models/mimo_v2_flash_nextn.py
sglang/srt/models/mindspore.py
sglang/srt/models/minicpm.py
sglang/srt/models/minicpm3.py
sglang/srt/models/minicpmo.py
sglang/srt/models/minicpmv.py
sglang/srt/models/minimax_m2.py
sglang/srt/models/ministral3.py
sglang/srt/models/mistral.py
sglang/srt/models/mistral_large_3.py
sglang/srt/models/mistral_large_3_eagle.py
sglang/srt/models/mixtral.py
sglang/srt/models/mixtral_quant.py
sglang/srt/models/mllama.py
sglang/srt/models/mllama4.py
sglang/srt/models/nano_nemotron_vl.py
sglang/srt/models/nemotron_h.py
sglang/srt/models/nemotron_nas.py
sglang/srt/models/nvila.py
sglang/srt/models/nvila_lite.py
sglang/srt/models/olmo.py
sglang/srt/models/olmo2.py
sglang/srt/models/olmoe.py
sglang/srt/models/opt.py
sglang/srt/models/orion.py
sglang/srt/models/paddleocr_vl.py
sglang/srt/models/persimmon.py
sglang/srt/models/phi.py
sglang/srt/models/phi3_small.py
sglang/srt/models/phi4mm.py
sglang/srt/models/phi4mm_audio.py
sglang/srt/models/phi4mm_utils.py
sglang/srt/models/phimoe.py
sglang/srt/models/pixtral.py
sglang/srt/models/points_v15_chat.py
sglang/srt/models/qwen.py
sglang/srt/models/qwen2.py
sglang/srt/models/qwen2_5_vl.py
sglang/srt/models/qwen2_audio.py
sglang/srt/models/qwen2_classification.py
sglang/srt/models/qwen2_eagle.py
sglang/srt/models/qwen2_moe.py
sglang/srt/models/qwen2_rm.py
sglang/srt/models/qwen2_vl.py
sglang/srt/models/qwen3.py
sglang/srt/models/qwen3_classification.py
sglang/srt/models/qwen3_moe.py
sglang/srt/models/qwen3_next.py
sglang/srt/models/qwen3_next_mtp.py
sglang/srt/models/qwen3_omni_moe.py
sglang/srt/models/qwen3_vl.py
sglang/srt/models/qwen3_vl_moe.py
sglang/srt/models/radio.py
sglang/srt/models/registry.py
sglang/srt/models/roberta.py
sglang/srt/models/sarashina2_vision.py
sglang/srt/models/siglip.py
sglang/srt/models/solar.py
sglang/srt/models/stablelm.py
sglang/srt/models/starcoder2.py
sglang/srt/models/step3_vl.py
sglang/srt/models/teleflm.py
sglang/srt/models/torch_native_llama.py
sglang/srt/models/transformers.py
sglang/srt/models/utils.py
sglang/srt/models/xverse.py
sglang/srt/models/xverse_moe.py
sglang/srt/models/yivl.py
sglang/srt/multimodal/customized_mm_processor_utils.py
sglang/srt/multimodal/internvl_utils.py
sglang/srt/multimodal/mm_utils.py
sglang/srt/multimodal/vit_cuda_graph_runner.py
sglang/srt/multimodal/processors/base_processor.py
sglang/srt/multimodal/processors/clip.py
sglang/srt/multimodal/processors/deepseek_ocr.py
sglang/srt/multimodal/processors/deepseek_vl_v2.py
sglang/srt/multimodal/processors/dots_vlm.py
sglang/srt/multimodal/processors/gemma3.py
sglang/srt/multimodal/processors/gemma3n.py
sglang/srt/multimodal/processors/glm4v.py
sglang/srt/multimodal/processors/glmasr.py
sglang/srt/multimodal/processors/internvl.py
sglang/srt/multimodal/processors/janus_pro.py
sglang/srt/multimodal/processors/kimi_vl.py
sglang/srt/multimodal/processors/llava.py
sglang/srt/multimodal/processors/midashenglm.py
sglang/srt/multimodal/processors/minicpm.py
sglang/srt/multimodal/processors/mlama.py
sglang/srt/multimodal/processors/mllama4.py
sglang/srt/multimodal/processors/nano_nemotron_vl.py
sglang/srt/multimodal/processors/nvila.py
sglang/srt/multimodal/processors/paddleocr_vlm.py
sglang/srt/multimodal/processors/phi4mm.py
sglang/srt/multimodal/processors/pixtral.py
sglang/srt/multimodal/processors/points_v15_chat.py
sglang/srt/multimodal/processors/qwen_audio.py
sglang/srt/multimodal/processors/qwen_vl.py
sglang/srt/multimodal/processors/sarashina2_vision.py
sglang/srt/multimodal/processors/step3_vl.py
sglang/srt/multiplex/multiplexing_mixin.py
sglang/srt/multiplex/pdmux_context.py
sglang/srt/parser/code_completion_parser.py
sglang/srt/parser/conversation.py
sglang/srt/parser/harmony_parser.py
sglang/srt/parser/jinja_template_utils.py
sglang/srt/parser/reasoning_parser.py
sglang/srt/sampling/custom_logit_processor.py
sglang/srt/sampling/sampling_batch_info.py
sglang/srt/sampling/sampling_params.py
sglang/srt/sampling/penaltylib/__init__.py
sglang/srt/sampling/penaltylib/frequency_penalty.py
sglang/srt/sampling/penaltylib/min_new_tokens.py
sglang/srt/sampling/penaltylib/orchestrator.py
sglang/srt/sampling/penaltylib/presence_penalty.py
sglang/srt/speculative/base_spec_worker.py
sglang/srt/speculative/draft_utils.py
sglang/srt/speculative/eagle_draft_cuda_graph_runner.py
sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py
sglang/srt/speculative/eagle_info.py
sglang/srt/speculative/eagle_info_v2.py
sglang/srt/speculative/eagle_utils.py
sglang/srt/speculative/eagle_worker.py
sglang/srt/speculative/eagle_worker_v2.py
sglang/srt/speculative/multi_layer_eagle_draft_extend_cuda_graph_runner.py
sglang/srt/speculative/multi_layer_eagle_utils.py
sglang/srt/speculative/multi_layer_eagle_worker.py
sglang/srt/speculative/multi_layer_eagle_worker_v2.py
sglang/srt/speculative/ngram_info.py
sglang/srt/speculative/ngram_worker.py
sglang/srt/speculative/spec_info.py
sglang/srt/speculative/spec_utils.py
sglang/srt/speculative/standalone_worker.py
sglang/srt/speculative/cpp_ngram/.clang-format
sglang/srt/speculative/cpp_ngram/ngram.cpp
sglang/srt/speculative/cpp_ngram/ngram.h
sglang/srt/speculative/cpp_ngram/ngram_cache.py
sglang/srt/speculative/cpp_ngram/ngram_cache_binding.cpp
sglang/srt/speculative/cpp_ngram/param.h
sglang/srt/speculative/cpp_ngram/queue.h
sglang/srt/tokenizer/tiktoken_tokenizer.py
sglang/srt/tracing/trace.py
sglang/srt/utils/__init__.py
sglang/srt/utils/aio_rwlock.py
sglang/srt/utils/bench_utils.py
sglang/srt/utils/common.py
sglang/srt/utils/cuda_ipc_transport_utils.py
sglang/srt/utils/device_timer.py
sglang/srt/utils/hf_transformers_utils.py
sglang/srt/utils/host_shared_memory.py
sglang/srt/utils/mistral_utils.py
sglang/srt/utils/multi_stream_utils.py
sglang/srt/utils/numa_utils.py
sglang/srt/utils/nvtx_pytorch_hooks.py
sglang/srt/utils/offloader.py
sglang/srt/utils/patch_torch.py
sglang/srt/utils/poll_based_barrier.py
sglang/srt/utils/profile_merger.py
sglang/srt/utils/profile_utils.py
sglang/srt/utils/request_logger.py
sglang/srt/utils/rpd_utils.py
sglang/srt/utils/slow_rank_detector.py
sglang/srt/utils/torch_memory_saver_adapter.py
sglang/srt/utils/watchdog.py
sglang/srt/utils/weight_checker.py
sglang/srt/weight_sync/tensor_bucket.py
sglang/srt/weight_sync/utils.py
sglang/test/__init__.py
sglang/test/accuracy_test_runner.py
sglang/test/doc_patch.py
sglang/test/few_shot_gsm8k.py
sglang/test/few_shot_gsm8k_engine.py
sglang/test/get_logits_ut.py
sglang/test/kl_test_utils.py
sglang/test/long_prompt.txt
sglang/test/lora_utils.py
sglang/test/nightly_bench_utils.py
sglang/test/nightly_utils.py
sglang/test/performance_test_runner.py
sglang/test/run_combined_tests.py
sglang/test/run_eval.py
sglang/test/runners.py
sglang/test/send_one.py
sglang/test/simple_eval_aime25.py
sglang/test/simple_eval_common.py
sglang/test/simple_eval_gpqa.py
sglang/test/simple_eval_humaneval.py
sglang/test/simple_eval_longbench_v2.py
sglang/test/simple_eval_math.py
sglang/test/simple_eval_mgsm.py
sglang/test/simple_eval_mmlu.py
sglang/test/simple_eval_mmmu_vlm.py
sglang/test/test_activation.py
sglang/test/test_block_fp8.py
sglang/test/test_block_fp8_deep_gemm_blackwell.py
sglang/test/test_custom_ops.py
sglang/test/test_cutlass_moe.py
sglang/test/test_cutlass_w16a16_moe.py
sglang/test/test_cutlass_w4a8_moe.py
sglang/test/test_deepep_utils.py
sglang/test/test_deterministic.py
sglang/test/test_deterministic_utils.py
sglang/test/test_dynamic_grad_mode.py
sglang/test/test_kvfp4_quant_dequant.py
sglang/test/test_layernorm.py
sglang/test/test_marlin_utils.py
sglang/test/test_programs.py
sglang/test/test_utils.py
sglang/test/ascend/__init__.py
sglang/test/ascend/gsm8k_ascend_mixin.py
sglang/test/ascend/vlm_utils.py
sglang/test/attention/__init__.py
sglang/test/attention/test_flashattn_backend.py
sglang/test/attention/test_flashattn_mla_backend.py
sglang/test/attention/test_prefix_chunk_info.py
sglang/test/attention/test_trtllm_mla_backend.py
sglang/test/ci/__init__.py
sglang/test/ci/ci_register.py
sglang/test/ci/ci_stress_utils.py
sglang/test/ci/ci_utils.py
sglang/test/ci/run_with_retry.py
sglang/test/kits/ebnf_constrained_kit.py
sglang/test/kits/gsm8k_accuracy_kit.py
sglang/test/kits/json_constrained_kit.py
sglang/test/kits/matched_stop_kit.py
sglang/test/kits/mmmu_vlm_kit.py
sglang/test/kits/radix_cache_server_kit.py
sglang/test/kits/regex_constrained_kit.py
sglang/test/kits/spec_decoding_kit.py
sglang/test/longbench_v2/__init__.py
sglang/test/longbench_v2/longbench_v2_evaluation.md
sglang/test/longbench_v2/test_longbench_v2_eval.py
sglang/test/longbench_v2/validate_longbench_v2.py
sglang/test/longbench_v2/validate_longbench_v2_standalone.py
sglang/test/server_fixtures/default_fixture.py
sglang/test/server_fixtures/disaggregation_fixture.py
sglang/test/server_fixtures/eagle_fixture.py
sglang/test/server_fixtures/mmmu_fixture.py
sglang/test/speculative/test_spec_utils.py