sglang 0.4.6.post4__py3-none-any.whl → 0.4.7__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- sglang/bench_offline_throughput.py +16 -10
- sglang/bench_one_batch.py +5 -4
- sglang/bench_one_batch_server.py +86 -22
- sglang/bench_serving.py +197 -110
- sglang/compile_deep_gemm.py +4 -4
- sglang/lang/backend/runtime_endpoint.py +24 -1
- sglang/profiler.py +167 -0
- sglang/srt/_custom_ops.py +34 -0
- sglang/srt/configs/internvl.py +8 -12
- sglang/srt/configs/model_config.py +66 -29
- sglang/srt/constrained/base_grammar_backend.py +5 -2
- sglang/srt/constrained/llguidance_backend.py +9 -8
- sglang/srt/constrained/outlines_backend.py +5 -4
- sglang/srt/constrained/xgrammar_backend.py +18 -18
- sglang/srt/conversation.py +47 -9
- sglang/srt/custom_op.py +38 -3
- sglang/srt/debug_utils.py +74 -0
- sglang/srt/disaggregation/common/__init__.py +1 -0
- sglang/srt/disaggregation/common/conn.py +407 -0
- sglang/srt/disaggregation/decode.py +187 -134
- sglang/srt/disaggregation/decode_schedule_batch_mixin.py +142 -0
- sglang/srt/disaggregation/fake/conn.py +4 -13
- sglang/srt/disaggregation/kv_events.py +412 -0
- sglang/srt/disaggregation/launch_lb.py +140 -0
- sglang/srt/disaggregation/mini_lb.py +84 -70
- sglang/srt/disaggregation/mooncake/conn.py +441 -140
- sglang/srt/disaggregation/mooncake/transfer_engine.py +31 -14
- sglang/srt/disaggregation/nixl/conn.py +124 -442
- sglang/srt/disaggregation/prefill.py +128 -44
- sglang/srt/disaggregation/utils.py +154 -6
- sglang/srt/distributed/device_communicators/pymscclpp.py +315 -0
- sglang/srt/distributed/parallel_state.py +52 -5
- sglang/srt/distributed/utils.py +3 -3
- sglang/srt/entrypoints/EngineBase.py +11 -0
- sglang/srt/entrypoints/engine.py +129 -12
- sglang/srt/entrypoints/http_server.py +21 -6
- sglang/srt/entrypoints/http_server_engine.py +5 -2
- sglang/srt/function_call/base_format_detector.py +302 -0
- sglang/srt/function_call/core_types.py +34 -0
- sglang/srt/function_call/deepseekv3_detector.py +205 -0
- sglang/srt/function_call/ebnf_composer.py +248 -0
- sglang/srt/function_call/function_call_parser.py +202 -0
- sglang/srt/function_call/llama32_detector.py +93 -0
- sglang/srt/function_call/mistral_detector.py +131 -0
- sglang/srt/function_call/pythonic_detector.py +229 -0
- sglang/srt/function_call/qwen25_detector.py +121 -0
- sglang/srt/function_call/utils.py +52 -0
- sglang/srt/hf_transformers_utils.py +50 -7
- sglang/srt/layers/attention/aiter_backend.py +878 -0
- sglang/srt/layers/attention/base_attn_backend.py +4 -0
- sglang/srt/layers/attention/cutlass_mla_backend.py +2 -19
- sglang/srt/layers/attention/flashattention_backend.py +166 -35
- sglang/srt/layers/attention/flashinfer_backend.py +45 -1
- sglang/srt/layers/attention/flashinfer_mla_backend.py +45 -5
- sglang/srt/layers/attention/flashmla_backend.py +340 -78
- sglang/srt/layers/attention/intel_amx_backend.py +128 -0
- sglang/srt/layers/attention/tbo_backend.py +232 -0
- sglang/srt/layers/attention/torch_native_backend.py +3 -0
- sglang/srt/layers/attention/triton_backend.py +247 -5
- sglang/srt/layers/attention/triton_ops/extend_attention.py +12 -4
- sglang/srt/layers/attention/utils.py +2 -2
- sglang/srt/layers/attention/vision.py +1 -1
- sglang/srt/layers/communicator.py +517 -0
- sglang/srt/layers/dp_attention.py +6 -15
- sglang/srt/layers/layernorm.py +30 -19
- sglang/srt/layers/moe/cutlass_moe.py +370 -0
- sglang/srt/layers/moe/cutlass_moe_params.py +169 -0
- sglang/srt/layers/moe/ep_moe/kernels.py +60 -17
- sglang/srt/layers/moe/ep_moe/layer.py +195 -87
- sglang/srt/layers/moe/ep_moe/token_dispatcher.py +88 -8
- sglang/srt/layers/moe/fused_moe_native.py +4 -0
- 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 +146 -0
- 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 +146 -0
- 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 +146 -0
- 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 +146 -0
- 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 +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_2_0/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- 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 +146 -0
- sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_3_1/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +220 -25
- sglang/srt/layers/moe/fused_moe_triton/layer.py +48 -4
- sglang/srt/layers/moe/topk.py +107 -24
- sglang/srt/layers/multimodal.py +70 -0
- sglang/srt/layers/quantization/__init__.py +10 -4
- sglang/srt/layers/quantization/blockwise_int8.py +3 -0
- sglang/srt/layers/quantization/compressed_tensors/compressed_tensors_moe.py +5 -0
- sglang/srt/layers/quantization/deep_gemm.py +60 -59
- sglang/srt/layers/quantization/fp8.py +113 -18
- sglang/srt/layers/quantization/fp8_kernel.py +118 -66
- sglang/srt/layers/quantization/fp8_utils.py +165 -43
- sglang/srt/layers/quantization/gptq.py +298 -6
- sglang/srt/layers/quantization/int8_kernel.py +18 -5
- sglang/srt/layers/quantization/modelopt_quant.py +334 -7
- sglang/srt/layers/quantization/moe_wna16.py +3 -0
- sglang/srt/layers/quantization/qoq.py +244 -0
- sglang/srt/layers/quantization/w8a8_fp8.py +3 -0
- sglang/srt/layers/quantization/w8a8_int8.py +3 -0
- sglang/srt/layers/rotary_embedding.py +6 -12
- sglang/srt/layers/sampler.py +80 -79
- sglang/srt/layers/utils.py +6 -0
- sglang/srt/lora/layers.py +12 -15
- sglang/srt/lora/lora.py +49 -5
- sglang/srt/lora/lora_manager.py +20 -8
- sglang/srt/lora/mem_pool.py +24 -16
- sglang/srt/lora/utils.py +17 -13
- sglang/srt/managers/data_parallel_controller.py +13 -5
- sglang/srt/managers/eplb_algorithms/__init__.py +63 -0
- sglang/srt/managers/eplb_algorithms/deepseek.py +223 -0
- sglang/srt/managers/eplb_algorithms/deepseek_vec.py +276 -0
- sglang/srt/managers/eplb_manager.py +96 -0
- sglang/srt/managers/expert_distribution.py +878 -56
- sglang/srt/managers/expert_location.py +448 -0
- sglang/srt/managers/expert_location_dispatch.py +108 -0
- sglang/srt/managers/io_struct.py +29 -5
- sglang/srt/managers/mm_utils.py +355 -151
- sglang/srt/managers/multimodal_processors/base_processor.py +299 -42
- sglang/srt/managers/multimodal_processors/deepseek_vl_v2.py +6 -1
- sglang/srt/managers/multimodal_processors/gemma3.py +15 -17
- sglang/srt/managers/multimodal_processors/internvl.py +18 -5
- sglang/srt/managers/multimodal_processors/janus_pro.py +7 -1
- sglang/srt/managers/multimodal_processors/kimi_vl.py +14 -32
- sglang/srt/managers/multimodal_processors/llava.py +3 -3
- sglang/srt/managers/multimodal_processors/minicpm.py +27 -32
- sglang/srt/managers/multimodal_processors/mllama4.py +6 -0
- sglang/srt/managers/multimodal_processors/phi4mm.py +87 -0
- sglang/srt/managers/multimodal_processors/pixtral.py +9 -9
- sglang/srt/managers/multimodal_processors/qwen_vl.py +35 -35
- sglang/srt/managers/schedule_batch.py +185 -55
- sglang/srt/managers/schedule_policy.py +4 -5
- sglang/srt/managers/scheduler.py +389 -154
- sglang/srt/managers/session_controller.py +1 -1
- sglang/srt/managers/tokenizer_manager.py +231 -39
- sglang/srt/managers/utils.py +0 -4
- sglang/srt/mem_cache/base_prefix_cache.py +3 -0
- sglang/srt/mem_cache/chunk_cache.py +3 -1
- sglang/srt/mem_cache/hiradix_cache.py +4 -4
- sglang/srt/mem_cache/memory_pool.py +74 -52
- sglang/srt/mem_cache/multimodal_cache.py +45 -0
- sglang/srt/mem_cache/radix_cache.py +58 -5
- sglang/srt/metrics/collector.py +11 -2
- sglang/srt/mm_utils.py +10 -0
- sglang/srt/model_executor/cuda_graph_runner.py +87 -65
- sglang/srt/model_executor/expert_location_updater.py +557 -0
- sglang/srt/model_executor/forward_batch_info.py +39 -14
- sglang/srt/model_executor/model_runner.py +231 -101
- sglang/srt/model_loader/loader.py +10 -6
- sglang/srt/model_loader/utils.py +67 -1
- sglang/srt/models/clip.py +5 -1
- sglang/srt/models/deepseek_nextn.py +1 -1
- sglang/srt/models/deepseek_v2.py +732 -403
- sglang/srt/models/exaone.py +8 -3
- sglang/srt/models/gemma3_causal.py +7 -0
- sglang/srt/models/gemma3_mm.py +75 -33
- sglang/srt/models/idefics2.py +342 -0
- sglang/srt/models/kimi_vl.py +4 -4
- sglang/srt/models/llama.py +1 -1
- sglang/srt/models/llama4.py +10 -2
- sglang/srt/models/llava.py +26 -18
- sglang/srt/models/mimo_mtp.py +220 -0
- sglang/srt/models/minicpmo.py +7 -17
- sglang/srt/models/minicpmv.py +3 -295
- sglang/srt/models/mistral.py +71 -1
- sglang/srt/models/mllama.py +3 -3
- sglang/srt/models/phi4mm.py +512 -0
- sglang/srt/models/qwen2.py +133 -35
- sglang/srt/models/qwen2_5_vl.py +5 -3
- sglang/srt/models/qwen2_eagle.py +4 -1
- sglang/srt/models/qwen2_moe.py +206 -69
- sglang/srt/models/qwen2_vl.py +3 -3
- sglang/srt/models/qwen3.py +92 -19
- sglang/srt/models/qwen3_moe.py +457 -55
- sglang/srt/models/registry.py +9 -1
- sglang/srt/models/siglip.py +294 -0
- sglang/srt/models/transformers.py +291 -0
- sglang/srt/openai_api/adapter.py +114 -40
- sglang/srt/openai_api/protocol.py +37 -2
- sglang/srt/openai_api/utils.py +172 -0
- sglang/srt/operations.py +189 -0
- sglang/srt/operations_strategy.py +207 -0
- sglang/srt/sampling/sampling_batch_info.py +13 -1
- sglang/srt/sampling/sampling_params.py +2 -1
- sglang/srt/server_args.py +235 -38
- sglang/srt/speculative/build_eagle_tree.py +8 -8
- sglang/srt/speculative/eagle_draft_cuda_graph_runner.py +8 -11
- sglang/srt/speculative/eagle_draft_extend_cuda_graph_runner.py +253 -0
- sglang/srt/speculative/eagle_utils.py +181 -90
- sglang/srt/speculative/eagle_worker.py +146 -21
- sglang/srt/two_batch_overlap.py +635 -0
- sglang/srt/utils.py +197 -19
- sglang/test/runners.py +16 -7
- sglang/test/send_one.py +4 -0
- sglang/test/test_cutlass_moe.py +278 -0
- sglang/test/test_fp4_moe.py +248 -0
- sglang/test/test_utils.py +81 -42
- sglang/utils.py +2 -2
- sglang/version.py +1 -1
- {sglang-0.4.6.post4.dist-info → sglang-0.4.7.dist-info}/METADATA +31 -19
- sglang-0.4.7.dist-info/RECORD +699 -0
- {sglang-0.4.6.post4.dist-info → sglang-0.4.7.dist-info}/WHEEL +1 -1
- sglang/srt/function_call_parser.py +0 -858
- sglang/srt/platforms/interface.py +0 -371
- sglang-0.4.6.post4.dist-info/RECORD +0 -646
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=3072,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=1,N=3072,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=3072,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json → triton_3_1_0/E=1,N=3072,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=3072,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=1,N=3072,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=144,N=512,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=144,N=512,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=16,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=1024,device_name=NVIDIA_H200.json → triton_3_1_0/E=16,N=1024,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=1344,device_name=NVIDIA_A100-SXM4-40GB.json → triton_3_1_0/E=16,N=1344,device_name=NVIDIA_A100-SXM4-40GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=1344,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=16,N=1344,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=1344,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=16,N=1344,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=16,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=16,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=16,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=2688,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=16,N=2688,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=2688,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=16,N=2688,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=3072,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=16,N=3072,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=3072,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json → triton_3_1_0/E=16,N=3072,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=3200,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=16,N=3200,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=3584,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=16,N=3584,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=16,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=6400,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=16,N=6400,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json → triton_3_1_0/E=16,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=16,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json → triton_3_1_0/E=16,N=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=16,N=800,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=16,N=800,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=160,N=192,device_name=NVIDIA_A800-SXM4-80GB.json → triton_3_1_0/E=160,N=192,device_name=NVIDIA_A800-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=20,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=20,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=24,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=24,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8.json → triton_3_1_0/E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json → triton_3_1_0/E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=128,device_name=NVIDIA_H20,block_shape=[128, 128].json → triton_3_1_0/E=256,N=128,device_name=NVIDIA_H20,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=128,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=128,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=256,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=256,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=256,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=256,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=256,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=256,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=256,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=256,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=256,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=64,device_name=NVIDIA_A800-SXM4-80GB.json → triton_3_1_0/E=256,N=64,device_name=NVIDIA_A800-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=64,device_name=NVIDIA_L20,dtype=int8_w8a8.json → triton_3_1_0/E=256,N=64,device_name=NVIDIA_L20,dtype=int8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=256,N=64,device_name=NVIDIA_L40S,dtype=int8_w8a8.json → triton_3_1_0/E=256,N=64,device_name=NVIDIA_L40S,dtype=int8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=1024,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=64,N=1024,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=1280,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=64,N=1280,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=1280,device_name=NVIDIA_A800-SXM4-80GB.json → triton_3_1_0/E=64,N=1280,device_name=NVIDIA_A800-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=1280,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=64,N=1280,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=1280,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=64,N=1280,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=1280,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=64,N=1280,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=1280,device_name=NVIDIA_H200.json → triton_3_1_0/E=64,N=1280,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=2560,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=64,N=2560,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=2560,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=64,N=2560,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=2560,device_name=NVIDIA_H200.json → triton_3_1_0/E=64,N=2560,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=320,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=64,N=320,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=320,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=64,N=320,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=320,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=64,N=320,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=320,device_name=NVIDIA_H200.json → triton_3_1_0/E=64,N=320,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_1_0/E=64,N=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=640,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=64,N=640,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=640,device_name=NVIDIA_A800-SXM4-80GB.json → triton_3_1_0/E=64,N=640,device_name=NVIDIA_A800-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=640,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json → triton_3_1_0/E=64,N=640,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=640,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=64,N=640,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=640,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=64,N=640,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=640,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=64,N=640,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=64,N=640,device_name=NVIDIA_H200.json → triton_3_1_0/E=64,N=640,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=14336,device_name=AMD_Instinct_MI300X.json → triton_3_1_0/E=8,N=14336,device_name=AMD_Instinct_MI300X.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=14336,device_name=AMD_Instinct_MI325X.json → triton_3_1_0/E=8,N=14336,device_name=AMD_Instinct_MI325X.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=14336,device_name=AMD_Radeon_Graphics.json → triton_3_1_0/E=8,N=14336,device_name=AMD_Radeon_Graphics.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=14336,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=14336,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=14336,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=14336,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=14336,device_name=NVIDIA_H200.json → triton_3_1_0/E=8,N=14336,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=1792,device_name=AMD_Instinct_MI300X.json → triton_3_1_0/E=8,N=1792,device_name=AMD_Instinct_MI300X.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=1792,device_name=AMD_Instinct_MI325X.json → triton_3_1_0/E=8,N=1792,device_name=AMD_Instinct_MI325X.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=1792,device_name=AMD_Radeon_Graphics.json → triton_3_1_0/E=8,N=1792,device_name=AMD_Radeon_Graphics.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=1792,device_name=NVIDIA_A100-SXM4-40GB.json → triton_3_1_0/E=8,N=1792,device_name=NVIDIA_A100-SXM4-40GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=8,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=1792,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=8,N=1792,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=1792,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=1792,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=1792,device_name=NVIDIA_H200.json → triton_3_1_0/E=8,N=1792,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=2048,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=8,N=2048,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=2048,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=2048,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=8,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=2048,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=2048,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=2048,device_name=NVIDIA_H200.json → triton_3_1_0/E=8,N=2048,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=AMD_Instinct_MI300X.json → triton_3_1_0/E=8,N=3584,device_name=AMD_Instinct_MI300X.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=AMD_Instinct_MI325X.json → triton_3_1_0/E=8,N=3584,device_name=AMD_Instinct_MI325X.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=AMD_Radeon_Graphics.json → triton_3_1_0/E=8,N=3584,device_name=AMD_Radeon_Graphics.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=NVIDIA_A100-SXM4-40GB.json → triton_3_1_0/E=8,N=3584,device_name=NVIDIA_A100-SXM4-40GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=8,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=3584,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=3584,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=8,N=3584,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=3584,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=NVIDIA_H200.json → triton_3_1_0/E=8,N=3584,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=3584,device_name=NVIDIA_L40S.json → triton_3_1_0/E=8,N=3584,device_name=NVIDIA_L40S.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=4096,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=4096,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=4096,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=4096,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=4096,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=4096,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=4096,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=8,N=4096,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=4096,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=4096,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=4096,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=8,N=4096,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=4096,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=4096,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=4096,device_name=NVIDIA_H200.json → triton_3_1_0/E=8,N=4096,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=7168,device_name=AMD_Instinct_MI300X.json → triton_3_1_0/E=8,N=7168,device_name=AMD_Instinct_MI300X.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=7168,device_name=AMD_Instinct_MI325X.json → triton_3_1_0/E=8,N=7168,device_name=AMD_Instinct_MI325X.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=7168,device_name=AMD_Radeon_Graphics.json → triton_3_1_0/E=8,N=7168,device_name=AMD_Radeon_Graphics.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json → triton_3_1_0/E=8,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_1_0/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=7168,device_name=NVIDIA_H200.json → triton_3_1_0/E=8,N=7168,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=8192,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=8192,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=8192,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=8192,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=8192,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=8192,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=8192,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=8192,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=8,N=8192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json → triton_3_1_0/E=8,N=8192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=192,device_name=NVIDIA_A800-SXM4-80GB.json → triton_3_2_0/E=128,N=192,device_name=NVIDIA_A800-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=192,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_2_0/E=128,N=192,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=192,device_name=NVIDIA_H20.json → triton_3_2_0/E=128,N=192,device_name=NVIDIA_H20.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=192,device_name=NVIDIA_H200.json → triton_3_2_0/E=128,N=192,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=384,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_2_0/E=128,N=384,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_2_0/E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=384,device_name=NVIDIA_H20.json → triton_3_2_0/E=128,N=384,device_name=NVIDIA_H20.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_2_0/E=128,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=384,device_name=NVIDIA_H200.json → triton_3_2_0/E=128,N=384,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=512,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_2_0/E=128,N=512,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=768,device_name=NVIDIA_A800-SXM4-80GB.json → triton_3_2_0/E=128,N=768,device_name=NVIDIA_A800-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=768,device_name=NVIDIA_H100_80GB_HBM3.json → triton_3_2_0/E=128,N=768,device_name=NVIDIA_H100_80GB_HBM3.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_2_0/E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=768,device_name=NVIDIA_H20.json → triton_3_2_0/E=128,N=768,device_name=NVIDIA_H20.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=768,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_2_0/E=128,N=768,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=768,device_name=NVIDIA_H200.json → triton_3_2_0/E=128,N=768,device_name=NVIDIA_H200.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=128,N=96,device_name=NVIDIA_H20.json → triton_3_2_0/E=128,N=96,device_name=NVIDIA_H20.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=264,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json → triton_3_2_0/E=264,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=264,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_2_0/E=264,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=264,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_2_0/E=264,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=264,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_2_0/E=264,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=272,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json → triton_3_2_0/E=272,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=272,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_2_0/E=272,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=272,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json → triton_3_2_0/E=272,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=272,N=64,device_name=NVIDIA_A800-SXM4-80GB.json → triton_3_2_0/E=272,N=64,device_name=NVIDIA_A800-SXM4-80GB.json} +0 -0
- /sglang/srt/layers/moe/fused_moe_triton/configs/{E=288,N=64,device_name=NVIDIA_A800-SXM4-80GB.json → triton_3_2_0/E=288,N=64,device_name=NVIDIA_A800-SXM4-80GB.json} +0 -0
- /sglang/srt/models/{xiaomi_mimo.py → mimo.py} +0 -0
- {sglang-0.4.6.post4.dist-info → sglang-0.4.7.dist-info}/licenses/LICENSE +0 -0
- {sglang-0.4.6.post4.dist-info → sglang-0.4.7.dist-info}/top_level.txt +0 -0