vllm-cpu 0.11.0.post2__cp312-cp312-manylinux_2_17_x86_64.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.
- vllm/_C.abi3.so +0 -0
- vllm/__init__.py +220 -0
- vllm/_bc_linter.py +59 -0
- vllm/_custom_ops.py +2044 -0
- vllm/_ipex_ops.py +393 -0
- vllm/_version.py +34 -0
- vllm/assets/__init__.py +0 -0
- vllm/assets/audio.py +45 -0
- vllm/assets/base.py +41 -0
- vllm/assets/image.py +50 -0
- vllm/assets/video.py +145 -0
- vllm/attention/__init__.py +15 -0
- vllm/attention/backends/__init__.py +0 -0
- vllm/attention/backends/abstract.py +204 -0
- vllm/attention/backends/utils.py +33 -0
- vllm/attention/layer.py +645 -0
- vllm/attention/layers/__init__.py +0 -0
- vllm/attention/layers/chunked_local_attention.py +93 -0
- vllm/attention/layers/cross_attention.py +162 -0
- vllm/attention/layers/encoder_only_attention.py +86 -0
- vllm/attention/ops/__init__.py +0 -0
- vllm/attention/ops/chunked_prefill_paged_decode.py +405 -0
- vllm/attention/ops/common.py +345 -0
- vllm/attention/ops/flashmla.py +192 -0
- vllm/attention/ops/merge_attn_states.py +43 -0
- vllm/attention/ops/paged_attn.py +262 -0
- vllm/attention/ops/pallas_kv_cache_update.py +124 -0
- vllm/attention/ops/prefix_prefill.py +928 -0
- vllm/attention/ops/rocm_aiter_mla.py +104 -0
- vllm/attention/ops/rocm_aiter_paged_attn.py +102 -0
- vllm/attention/ops/triton_decode_attention.py +691 -0
- vllm/attention/ops/triton_flash_attention.py +984 -0
- vllm/attention/ops/triton_merge_attn_states.py +97 -0
- vllm/attention/ops/triton_reshape_and_cache_flash.py +175 -0
- vllm/attention/ops/triton_unified_attention.py +894 -0
- vllm/attention/selector.py +245 -0
- vllm/attention/utils/__init__.py +0 -0
- vllm/attention/utils/fa_utils.py +85 -0
- vllm/attention/utils/kv_sharing_utils.py +33 -0
- vllm/beam_search.py +87 -0
- vllm/benchmarks/__init__.py +0 -0
- vllm/benchmarks/datasets.py +2723 -0
- vllm/benchmarks/latency.py +170 -0
- vllm/benchmarks/lib/__init__.py +3 -0
- vllm/benchmarks/lib/endpoint_request_func.py +533 -0
- vllm/benchmarks/lib/ready_checker.py +73 -0
- vllm/benchmarks/lib/utils.py +80 -0
- vllm/benchmarks/serve.py +1358 -0
- vllm/benchmarks/throughput.py +696 -0
- vllm/collect_env.py +823 -0
- vllm/compilation/__init__.py +0 -0
- vllm/compilation/activation_quant_fusion.py +189 -0
- vllm/compilation/backends.py +650 -0
- vllm/compilation/base_static_graph.py +56 -0
- vllm/compilation/collective_fusion.py +1188 -0
- vllm/compilation/compiler_interface.py +573 -0
- vllm/compilation/counter.py +47 -0
- vllm/compilation/cuda_graph.py +199 -0
- vllm/compilation/cuda_piecewise_backend.py +117 -0
- vllm/compilation/decorators.py +400 -0
- vllm/compilation/fix_functionalization.py +205 -0
- vllm/compilation/fusion.py +383 -0
- vllm/compilation/fusion_attn.py +295 -0
- vllm/compilation/fx_utils.py +84 -0
- vllm/compilation/inductor_pass.py +136 -0
- vllm/compilation/monitor.py +57 -0
- vllm/compilation/noop_elimination.py +158 -0
- vllm/compilation/pass_manager.py +125 -0
- vllm/compilation/post_cleanup.py +20 -0
- vllm/compilation/sequence_parallelism.py +478 -0
- vllm/compilation/torch25_custom_graph_pass.py +42 -0
- vllm/compilation/vllm_inductor_pass.py +156 -0
- vllm/compilation/wrapper.py +136 -0
- vllm/config/__init__.py +814 -0
- vllm/config/cache.py +220 -0
- vllm/config/compilation.py +673 -0
- vllm/config/device.py +74 -0
- vllm/config/kv_events.py +50 -0
- vllm/config/kv_transfer.py +111 -0
- vllm/config/load.py +113 -0
- vllm/config/lora.py +132 -0
- vllm/config/model.py +1912 -0
- vllm/config/multimodal.py +129 -0
- vllm/config/observability.py +99 -0
- vllm/config/parallel.py +524 -0
- vllm/config/pooler.py +97 -0
- vllm/config/scheduler.py +287 -0
- vllm/config/speculative.py +568 -0
- vllm/config/speech_to_text.py +39 -0
- vllm/config/structured_outputs.py +64 -0
- vllm/config/utils.py +145 -0
- vllm/connections.py +186 -0
- vllm/device_allocator/__init__.py +0 -0
- vllm/device_allocator/cumem.py +311 -0
- vllm/distributed/__init__.py +6 -0
- vllm/distributed/communication_op.py +41 -0
- vllm/distributed/device_communicators/__init__.py +0 -0
- vllm/distributed/device_communicators/all2all.py +440 -0
- vllm/distributed/device_communicators/all_reduce_utils.py +317 -0
- vllm/distributed/device_communicators/base_device_communicator.py +295 -0
- vllm/distributed/device_communicators/cpu_communicator.py +201 -0
- vllm/distributed/device_communicators/cuda_communicator.py +323 -0
- vllm/distributed/device_communicators/cuda_wrapper.py +180 -0
- vllm/distributed/device_communicators/custom_all_reduce.py +311 -0
- vllm/distributed/device_communicators/mnnvl_compat.py +28 -0
- vllm/distributed/device_communicators/pynccl.py +340 -0
- vllm/distributed/device_communicators/pynccl_allocator.py +186 -0
- vllm/distributed/device_communicators/pynccl_wrapper.py +416 -0
- vllm/distributed/device_communicators/quick_all_reduce.py +278 -0
- vllm/distributed/device_communicators/ray_communicator.py +258 -0
- vllm/distributed/device_communicators/shm_broadcast.py +589 -0
- vllm/distributed/device_communicators/shm_object_storage.py +635 -0
- vllm/distributed/device_communicators/symm_mem.py +136 -0
- vllm/distributed/device_communicators/tpu_communicator.py +102 -0
- vllm/distributed/device_communicators/xpu_communicator.py +94 -0
- vllm/distributed/eplb/__init__.py +8 -0
- vllm/distributed/eplb/eplb_state.py +620 -0
- vllm/distributed/eplb/rebalance_algo.py +239 -0
- vllm/distributed/eplb/rebalance_execute.py +424 -0
- vllm/distributed/kv_events.py +362 -0
- vllm/distributed/kv_transfer/README.md +29 -0
- vllm/distributed/kv_transfer/__init__.py +13 -0
- vllm/distributed/kv_transfer/disagg_prefill_workflow.jpg +0 -0
- vllm/distributed/kv_transfer/kv_connector/__init__.py +0 -0
- vllm/distributed/kv_transfer/kv_connector/base.py +10 -0
- vllm/distributed/kv_transfer/kv_connector/factory.py +113 -0
- vllm/distributed/kv_transfer/kv_connector/utils.py +261 -0
- vllm/distributed/kv_transfer/kv_connector/v1/__init__.py +6 -0
- vllm/distributed/kv_transfer/kv_connector/v1/base.py +388 -0
- vllm/distributed/kv_transfer/kv_connector/v1/lmcache_connector.py +168 -0
- vllm/distributed/kv_transfer/kv_connector/v1/metrics.py +100 -0
- vllm/distributed/kv_transfer/kv_connector/v1/multi_connector.py +328 -0
- vllm/distributed/kv_transfer/kv_connector/v1/nixl_connector.py +1473 -0
- vllm/distributed/kv_transfer/kv_connector/v1/offloading_connector.py +485 -0
- vllm/distributed/kv_transfer/kv_connector/v1/p2p/__init__.py +0 -0
- vllm/distributed/kv_transfer/kv_connector/v1/p2p/p2p_nccl_connector.py +488 -0
- vllm/distributed/kv_transfer/kv_connector/v1/p2p/p2p_nccl_engine.py +550 -0
- vllm/distributed/kv_transfer/kv_connector/v1/p2p/tensor_memory_pool.py +267 -0
- vllm/distributed/kv_transfer/kv_connector/v1/shared_storage_connector.py +418 -0
- vllm/distributed/kv_transfer/kv_lookup_buffer/__init__.py +0 -0
- vllm/distributed/kv_transfer/kv_lookup_buffer/base.py +175 -0
- vllm/distributed/kv_transfer/kv_lookup_buffer/mooncake_store.py +161 -0
- vllm/distributed/kv_transfer/kv_lookup_buffer/simple_buffer.py +237 -0
- vllm/distributed/kv_transfer/kv_pipe/__init__.py +0 -0
- vllm/distributed/kv_transfer/kv_pipe/base.py +67 -0
- vllm/distributed/kv_transfer/kv_pipe/mooncake_pipe.py +290 -0
- vllm/distributed/kv_transfer/kv_pipe/pynccl_pipe.py +280 -0
- vllm/distributed/kv_transfer/kv_transfer_state.py +73 -0
- vllm/distributed/parallel_state.py +1532 -0
- vllm/distributed/tpu_distributed_utils.py +178 -0
- vllm/distributed/utils.py +536 -0
- vllm/engine/__init__.py +0 -0
- vllm/engine/arg_utils.py +1778 -0
- vllm/engine/async_llm_engine.py +6 -0
- vllm/engine/llm_engine.py +6 -0
- vllm/engine/metrics.py +577 -0
- vllm/engine/metrics_types.py +84 -0
- vllm/engine/protocol.py +333 -0
- vllm/entrypoints/__init__.py +0 -0
- vllm/entrypoints/api_server.py +178 -0
- vllm/entrypoints/chat_utils.py +1705 -0
- vllm/entrypoints/cli/__init__.py +12 -0
- vllm/entrypoints/cli/benchmark/__init__.py +0 -0
- vllm/entrypoints/cli/benchmark/base.py +25 -0
- vllm/entrypoints/cli/benchmark/latency.py +21 -0
- vllm/entrypoints/cli/benchmark/main.py +55 -0
- vllm/entrypoints/cli/benchmark/serve.py +21 -0
- vllm/entrypoints/cli/benchmark/throughput.py +21 -0
- vllm/entrypoints/cli/collect_env.py +36 -0
- vllm/entrypoints/cli/main.py +60 -0
- vllm/entrypoints/cli/openai.py +233 -0
- vllm/entrypoints/cli/run_batch.py +67 -0
- vllm/entrypoints/cli/serve.py +232 -0
- vllm/entrypoints/cli/types.py +29 -0
- vllm/entrypoints/constants.py +10 -0
- vllm/entrypoints/context.py +481 -0
- vllm/entrypoints/harmony_utils.py +436 -0
- vllm/entrypoints/launcher.py +164 -0
- vllm/entrypoints/llm.py +1629 -0
- vllm/entrypoints/logger.py +79 -0
- vllm/entrypoints/openai/__init__.py +0 -0
- vllm/entrypoints/openai/api_server.py +1953 -0
- vllm/entrypoints/openai/cli_args.py +288 -0
- vllm/entrypoints/openai/logits_processors.py +90 -0
- vllm/entrypoints/openai/protocol.py +2757 -0
- vllm/entrypoints/openai/run_batch.py +491 -0
- vllm/entrypoints/openai/serving_chat.py +1597 -0
- vllm/entrypoints/openai/serving_classification.py +173 -0
- vllm/entrypoints/openai/serving_completion.py +692 -0
- vllm/entrypoints/openai/serving_embedding.py +631 -0
- vllm/entrypoints/openai/serving_engine.py +992 -0
- vllm/entrypoints/openai/serving_models.py +288 -0
- vllm/entrypoints/openai/serving_pooling.py +276 -0
- vllm/entrypoints/openai/serving_responses.py +1709 -0
- vllm/entrypoints/openai/serving_score.py +479 -0
- vllm/entrypoints/openai/serving_tokenization.py +196 -0
- vllm/entrypoints/openai/serving_transcription.py +136 -0
- vllm/entrypoints/openai/speech_to_text.py +388 -0
- vllm/entrypoints/openai/tool_parsers/__init__.py +55 -0
- vllm/entrypoints/openai/tool_parsers/abstract_tool_parser.py +164 -0
- vllm/entrypoints/openai/tool_parsers/deepseekv31_tool_parser.py +367 -0
- vllm/entrypoints/openai/tool_parsers/deepseekv3_tool_parser.py +370 -0
- vllm/entrypoints/openai/tool_parsers/glm4_moe_tool_parser.py +185 -0
- vllm/entrypoints/openai/tool_parsers/granite_20b_fc_tool_parser.py +259 -0
- vllm/entrypoints/openai/tool_parsers/granite_tool_parser.py +237 -0
- vllm/entrypoints/openai/tool_parsers/hermes_tool_parser.py +455 -0
- vllm/entrypoints/openai/tool_parsers/hunyuan_a13b_tool_parser.py +372 -0
- vllm/entrypoints/openai/tool_parsers/internlm2_tool_parser.py +216 -0
- vllm/entrypoints/openai/tool_parsers/jamba_tool_parser.py +308 -0
- vllm/entrypoints/openai/tool_parsers/kimi_k2_tool_parser.py +377 -0
- vllm/entrypoints/openai/tool_parsers/llama4_pythonic_tool_parser.py +316 -0
- vllm/entrypoints/openai/tool_parsers/llama_tool_parser.py +269 -0
- vllm/entrypoints/openai/tool_parsers/longcat_tool_parser.py +39 -0
- vllm/entrypoints/openai/tool_parsers/minimax_tool_parser.py +816 -0
- vllm/entrypoints/openai/tool_parsers/mistral_tool_parser.py +369 -0
- vllm/entrypoints/openai/tool_parsers/openai_tool_parser.py +93 -0
- vllm/entrypoints/openai/tool_parsers/phi4mini_tool_parser.py +112 -0
- vllm/entrypoints/openai/tool_parsers/pythonic_tool_parser.py +308 -0
- vllm/entrypoints/openai/tool_parsers/qwen3coder_tool_parser.py +707 -0
- vllm/entrypoints/openai/tool_parsers/qwen3xml_tool_parser.py +1137 -0
- vllm/entrypoints/openai/tool_parsers/seed_oss_tool_parser.py +679 -0
- vllm/entrypoints/openai/tool_parsers/step3_tool_parser.py +296 -0
- vllm/entrypoints/openai/tool_parsers/utils.py +124 -0
- vllm/entrypoints/openai/tool_parsers/xlam_tool_parser.py +524 -0
- vllm/entrypoints/renderer.py +395 -0
- vllm/entrypoints/score_utils.py +232 -0
- vllm/entrypoints/ssl.py +75 -0
- vllm/entrypoints/tool.py +139 -0
- vllm/entrypoints/tool_server.py +206 -0
- vllm/entrypoints/utils.py +233 -0
- vllm/env_override.py +23 -0
- vllm/envs.py +1590 -0
- vllm/executor/__init__.py +0 -0
- vllm/executor/executor_base.py +381 -0
- vllm/executor/msgspec_utils.py +35 -0
- vllm/executor/ray_distributed_executor.py +699 -0
- vllm/executor/ray_utils.py +410 -0
- vllm/executor/uniproc_executor.py +176 -0
- vllm/forward_context.py +402 -0
- vllm/inputs/__init__.py +30 -0
- vllm/inputs/data.py +356 -0
- vllm/inputs/parse.py +151 -0
- vllm/inputs/preprocess.py +664 -0
- vllm/logger.py +229 -0
- vllm/logging_utils/__init__.py +10 -0
- vllm/logging_utils/dump_input.py +81 -0
- vllm/logging_utils/formatter.py +79 -0
- vllm/logging_utils/log_time.py +32 -0
- vllm/logits_process.py +119 -0
- vllm/logprobs.py +28 -0
- vllm/lora/__init__.py +0 -0
- vllm/lora/layers/__init__.py +34 -0
- vllm/lora/layers/base.py +69 -0
- vllm/lora/layers/base_linear.py +185 -0
- vllm/lora/layers/column_parallel_linear.py +609 -0
- vllm/lora/layers/logits_processor.py +247 -0
- vllm/lora/layers/qkv_x_parallel_linear.py +8 -0
- vllm/lora/layers/replicated_linear.py +60 -0
- vllm/lora/layers/row_parallel_linear.py +196 -0
- vllm/lora/layers/utils.py +65 -0
- vllm/lora/layers/vocal_parallel_embedding.py +174 -0
- vllm/lora/lora_weights.py +199 -0
- vllm/lora/models.py +816 -0
- vllm/lora/ops/__init__.py +0 -0
- vllm/lora/ops/ipex_ops/__init__.py +7 -0
- vllm/lora/ops/ipex_ops/lora_ops.py +44 -0
- vllm/lora/ops/torch_ops/__init__.py +16 -0
- vllm/lora/ops/torch_ops/lora_ops.py +119 -0
- vllm/lora/ops/triton_ops/__init__.py +12 -0
- vllm/lora/ops/triton_ops/kernel_utils.py +243 -0
- vllm/lora/ops/triton_ops/lora_expand_op.py +289 -0
- vllm/lora/ops/triton_ops/lora_kernel_metadata.py +148 -0
- vllm/lora/ops/triton_ops/lora_shrink_op.py +243 -0
- vllm/lora/ops/triton_ops/utils.py +126 -0
- vllm/lora/ops/xla_ops/__init__.py +7 -0
- vllm/lora/ops/xla_ops/lora_ops.py +144 -0
- vllm/lora/peft_helper.py +127 -0
- vllm/lora/punica_wrapper/__init__.py +10 -0
- vllm/lora/punica_wrapper/punica_base.py +458 -0
- vllm/lora/punica_wrapper/punica_cpu.py +349 -0
- vllm/lora/punica_wrapper/punica_gpu.py +272 -0
- vllm/lora/punica_wrapper/punica_selector.py +20 -0
- vllm/lora/punica_wrapper/punica_tpu.py +391 -0
- vllm/lora/punica_wrapper/punica_xpu.py +276 -0
- vllm/lora/punica_wrapper/utils.py +136 -0
- vllm/lora/request.py +97 -0
- vllm/lora/resolver.py +85 -0
- vllm/lora/utils.py +246 -0
- vllm/lora/worker_manager.py +267 -0
- vllm/model_executor/__init__.py +12 -0
- vllm/model_executor/custom_op.py +194 -0
- vllm/model_executor/layers/__init__.py +0 -0
- vllm/model_executor/layers/activation.py +575 -0
- vllm/model_executor/layers/attention_layer_base.py +23 -0
- vllm/model_executor/layers/fla/__init__.py +8 -0
- vllm/model_executor/layers/fla/ops/__init__.py +17 -0
- vllm/model_executor/layers/fla/ops/chunk.py +225 -0
- vllm/model_executor/layers/fla/ops/chunk_delta_h.py +290 -0
- vllm/model_executor/layers/fla/ops/chunk_o.py +177 -0
- vllm/model_executor/layers/fla/ops/chunk_scaled_dot_kkt.py +140 -0
- vllm/model_executor/layers/fla/ops/cumsum.py +226 -0
- vllm/model_executor/layers/fla/ops/fused_recurrent.py +366 -0
- vllm/model_executor/layers/fla/ops/index.py +39 -0
- vllm/model_executor/layers/fla/ops/l2norm.py +143 -0
- vllm/model_executor/layers/fla/ops/layernorm_guard.py +337 -0
- vllm/model_executor/layers/fla/ops/op.py +39 -0
- vllm/model_executor/layers/fla/ops/solve_tril.py +365 -0
- vllm/model_executor/layers/fla/ops/utils.py +180 -0
- vllm/model_executor/layers/fla/ops/wy_fast.py +114 -0
- vllm/model_executor/layers/fused_moe/__init__.py +89 -0
- vllm/model_executor/layers/fused_moe/batched_deep_gemm_moe.py +322 -0
- vllm/model_executor/layers/fused_moe/batched_triton_or_deep_gemm_moe.py +141 -0
- vllm/model_executor/layers/fused_moe/config.py +804 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=3072,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=3072,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=3072,device_name=NVIDIA_H100_80GB_HBM3.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=1024,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=1024,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=1024,device_name=NVIDIA_H100,dtype=fp8_w8a8.json +123 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=192,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=192,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=192,device_name=NVIDIA_H20-3e.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=192,device_name=NVIDIA_H20.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=192,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=352,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +122 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=384,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=384,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=384,device_name=NVIDIA_GB200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=384,device_name=NVIDIA_H20-3e,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=384,device_name=NVIDIA_H20-3e.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=384,device_name=NVIDIA_H20.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=384,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=512,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=704,device_name=NVIDIA_B200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=704,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +114 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=768,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=768,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=768,device_name=NVIDIA_GB200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=768,device_name=NVIDIA_H20-3e,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=768,device_name=NVIDIA_H20.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=768,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=768,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=128,N=96,device_name=NVIDIA_H20.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=1024,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=1024,device_name=NVIDIA_B200,dtype=fp8_w8a8.json +147 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=1024,device_name=NVIDIA_B200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=1024,device_name=NVIDIA_H100.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=1344,device_name=NVIDIA_A100-SXM4-40GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=1344,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=1344,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=2688,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=2688,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=3072,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=3072,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=3200,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +130 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=3584,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json +218 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=6400,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +130 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=int8_w8a16.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=16,N=800,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +130 -0
- vllm/model_executor/layers/fused_moe/configs/E=160,N=192,device_name=NVIDIA_A800-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=160,N=192,device_name=NVIDIA_H20-3e.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=160,N=320,device_name=NVIDIA_H20-3e.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=160,N=384,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=160,N=640,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=160,N=640,device_name=NVIDIA_GB200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=160,N=640,device_name=NVIDIA_H100,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=20,N=2560,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=20,N=2560,device_name=NVIDIA_GB200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=20,N=2560,device_name=NVIDIA_H100,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=20,N=2560,device_name=NVIDIA_H20-3e,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=1024,device_name=AMD_Instinct_MI325X,block_shape=[128,128].json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=1024,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=128,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=256,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=256,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=256,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=256,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=256,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=256,device_name=NVIDIA_H20-3e,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=256,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=512,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=512,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=256,N=64,device_name=NVIDIA_A800-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=384,N=128,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=384,N=128,device_name=NVIDIA_GB200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=384,N=128,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=384,N=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=384,N=256,device_name=NVIDIA_GB200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=40,N=2560,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=40,N=2560,device_name=NVIDIA_GB200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=40,N=2560,device_name=NVIDIA_H100,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=128,device_name=NVIDIA_B200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=128,device_name=NVIDIA_GB200,dtype=fp8_w8a8.json +147 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=128,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=128,device_name=NVIDIA_H20-3e.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=128,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=256,device_name=NVIDIA_B200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=256,device_name=NVIDIA_GB200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=256,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=256,device_name=NVIDIA_H20-3e.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=256,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=512,device_name=NVIDIA_B200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=512,device_name=NVIDIA_GB200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=512,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=512,device_name=NVIDIA_H20-3e.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=512,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=64,device_name=NVIDIA_B200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=64,device_name=NVIDIA_H20-3e.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=512,N=64,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=60,N=1408,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=60,N=176,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=60,N=352,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=60,N=704,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=62,N=128,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=62,N=256,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=62,N=256,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=62,N=512,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=62,N=512,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=1280,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=1280,device_name=NVIDIA_A800-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=1280,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=1280,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=1280,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=1280,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=1536,device_name=NVIDIA_H20,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=2560,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=2560,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=2560,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=3072,device_name=NVIDIA_H20,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=3072,device_name=NVIDIA_H20.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=320,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=320,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=320,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=320,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=384,device_name=NVIDIA_H20.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=640,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=640,device_name=NVIDIA_A800-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=640,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=640,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=640,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=640,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=640,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=768,device_name=NVIDIA_H20.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=64,N=896,device_name=NVIDIA_H20.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=72,N=192,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=72,N=384,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=72,N=384,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=72,N=768,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=72,N=768,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=14336,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=14336,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=14336,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=14336,device_name=AMD_Instinct_MI325X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=14336,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +138 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=14336,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=14336,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=16384,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=16384,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=16384,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=16384,device_name=AMD_Instinct_MI325X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=1792,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=1792,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=1792,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=1792,device_name=AMD_Instinct_MI325X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=1792,device_name=NVIDIA_A100-SXM4-40GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=1792,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=1792,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=1792,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=1792,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=AMD_Instinct_MI325X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +154 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=2048,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=AMD_Instinct_MI325X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=NVIDIA_A100-SXM4-40GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=NVIDIA_GeForce_RTX_4090,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=3584,device_name=NVIDIA_L40S.json +173 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=4096,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=4096,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=4096,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=4096,device_name=AMD_Instinct_MI325X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=4096,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=4096,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=4096,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=4096,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=4096,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=7168,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=7168,device_name=AMD_Instinct_MI325X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=7168,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=7168,device_name=NVIDIA_H200.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=8192,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=8192,device_name=AMD_Instinct_MI300X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=8192,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8.json +164 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=8192,device_name=AMD_Instinct_MI325X.json +200 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=8192,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/E=8,N=8192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json +146 -0
- vllm/model_executor/layers/fused_moe/configs/README +12 -0
- vllm/model_executor/layers/fused_moe/cpu_fused_moe.py +300 -0
- vllm/model_executor/layers/fused_moe/cutlass_moe.py +957 -0
- vllm/model_executor/layers/fused_moe/deep_gemm_moe.py +362 -0
- vllm/model_executor/layers/fused_moe/deep_gemm_utils.py +413 -0
- vllm/model_executor/layers/fused_moe/deepep_ht_prepare_finalize.py +361 -0
- vllm/model_executor/layers/fused_moe/deepep_ll_prepare_finalize.py +274 -0
- vllm/model_executor/layers/fused_moe/flashinfer_cutlass_moe.py +268 -0
- vllm/model_executor/layers/fused_moe/flashinfer_cutlass_prepare_finalize.py +300 -0
- vllm/model_executor/layers/fused_moe/flashinfer_trtllm_moe.py +184 -0
- vllm/model_executor/layers/fused_moe/fused_batched_moe.py +993 -0
- vllm/model_executor/layers/fused_moe/fused_marlin_moe.py +239 -0
- vllm/model_executor/layers/fused_moe/fused_moe.py +1890 -0
- vllm/model_executor/layers/fused_moe/gpt_oss_triton_kernels_moe.py +307 -0
- vllm/model_executor/layers/fused_moe/layer.py +2195 -0
- vllm/model_executor/layers/fused_moe/modular_kernel.py +1038 -0
- vllm/model_executor/layers/fused_moe/moe_align_block_size.py +87 -0
- vllm/model_executor/layers/fused_moe/moe_pallas.py +80 -0
- vllm/model_executor/layers/fused_moe/moe_permute_unpermute.py +205 -0
- vllm/model_executor/layers/fused_moe/moe_torch_iterative.py +60 -0
- vllm/model_executor/layers/fused_moe/pplx_prepare_finalize.py +341 -0
- vllm/model_executor/layers/fused_moe/prepare_finalize.py +70 -0
- vllm/model_executor/layers/fused_moe/rocm_aiter_fused_moe.py +424 -0
- vllm/model_executor/layers/fused_moe/routing_simulator.py +291 -0
- vllm/model_executor/layers/fused_moe/topk_weight_and_reduce.py +146 -0
- vllm/model_executor/layers/fused_moe/triton_deep_gemm_moe.py +143 -0
- vllm/model_executor/layers/fused_moe/trtllm_moe.py +191 -0
- vllm/model_executor/layers/fused_moe/utils.py +274 -0
- vllm/model_executor/layers/layernorm.py +395 -0
- vllm/model_executor/layers/lightning_attn.py +661 -0
- vllm/model_executor/layers/linear.py +1603 -0
- vllm/model_executor/layers/logits_processor.py +106 -0
- vllm/model_executor/layers/mamba/__init__.py +0 -0
- vllm/model_executor/layers/mamba/abstract.py +42 -0
- vllm/model_executor/layers/mamba/linear_attn.py +403 -0
- vllm/model_executor/layers/mamba/mamba_mixer.py +466 -0
- vllm/model_executor/layers/mamba/mamba_mixer2.py +764 -0
- vllm/model_executor/layers/mamba/mamba_utils.py +186 -0
- vllm/model_executor/layers/mamba/ops/__init__.py +0 -0
- vllm/model_executor/layers/mamba/ops/causal_conv1d.py +1092 -0
- vllm/model_executor/layers/mamba/ops/layernorm_gated.py +168 -0
- vllm/model_executor/layers/mamba/ops/mamba_ssm.py +414 -0
- vllm/model_executor/layers/mamba/ops/ssd_bmm.py +242 -0
- vllm/model_executor/layers/mamba/ops/ssd_chunk_scan.py +527 -0
- vllm/model_executor/layers/mamba/ops/ssd_chunk_state.py +724 -0
- vllm/model_executor/layers/mamba/ops/ssd_combined.py +238 -0
- vllm/model_executor/layers/mamba/ops/ssd_state_passing.py +200 -0
- vllm/model_executor/layers/mamba/short_conv.py +253 -0
- vllm/model_executor/layers/mla.py +173 -0
- vllm/model_executor/layers/pooler.py +719 -0
- vllm/model_executor/layers/quantization/__init__.py +157 -0
- vllm/model_executor/layers/quantization/auto_round.py +388 -0
- vllm/model_executor/layers/quantization/awq.py +228 -0
- vllm/model_executor/layers/quantization/awq_marlin.py +554 -0
- vllm/model_executor/layers/quantization/awq_triton.py +320 -0
- vllm/model_executor/layers/quantization/base_config.py +170 -0
- vllm/model_executor/layers/quantization/bitblas.py +464 -0
- vllm/model_executor/layers/quantization/bitsandbytes.py +627 -0
- vllm/model_executor/layers/quantization/compressed_tensors/__init__.py +0 -0
- vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py +797 -0
- vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors_moe.py +2074 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/__init__.py +27 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_24.py +366 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_scheme.py +55 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w4a16_24.py +160 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w4a16_nvfp4.py +105 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w4a4_nvfp4.py +185 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w4a8_fp8.py +169 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w4a8_int.py +135 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a16_fp8.py +121 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_fp8.py +157 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_int8.py +111 -0
- vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_wNa16.py +201 -0
- vllm/model_executor/layers/quantization/compressed_tensors/transform/__init__.py +0 -0
- vllm/model_executor/layers/quantization/compressed_tensors/transform/linear.py +238 -0
- vllm/model_executor/layers/quantization/compressed_tensors/transform/module.py +153 -0
- vllm/model_executor/layers/quantization/compressed_tensors/transform/schemes/__init__.py +0 -0
- vllm/model_executor/layers/quantization/compressed_tensors/transform/schemes/linear_qutlass_nvfp4.py +46 -0
- vllm/model_executor/layers/quantization/compressed_tensors/transform/utils.py +13 -0
- vllm/model_executor/layers/quantization/compressed_tensors/triton_scaled_mm.py +206 -0
- vllm/model_executor/layers/quantization/compressed_tensors/utils.py +216 -0
- vllm/model_executor/layers/quantization/deepspeedfp.py +196 -0
- vllm/model_executor/layers/quantization/experts_int8.py +223 -0
- vllm/model_executor/layers/quantization/fbgemm_fp8.py +172 -0
- vllm/model_executor/layers/quantization/fp8.py +1098 -0
- vllm/model_executor/layers/quantization/gguf.py +599 -0
- vllm/model_executor/layers/quantization/gptq.py +340 -0
- vllm/model_executor/layers/quantization/gptq_bitblas.py +448 -0
- vllm/model_executor/layers/quantization/gptq_marlin.py +751 -0
- vllm/model_executor/layers/quantization/gptq_marlin_24.py +297 -0
- vllm/model_executor/layers/quantization/hqq_marlin.py +333 -0
- vllm/model_executor/layers/quantization/inc.py +61 -0
- vllm/model_executor/layers/quantization/input_quant_fp8.py +156 -0
- vllm/model_executor/layers/quantization/ipex_quant.py +415 -0
- vllm/model_executor/layers/quantization/kernels/__init__.py +0 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/MPLinearKernel.py +91 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/__init__.py +93 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/allspark.py +116 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/bitblas.py +302 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/conch.py +92 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/cutlass.py +117 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/dynamic_4bit.py +92 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/exllama.py +143 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/machete.py +144 -0
- vllm/model_executor/layers/quantization/kernels/mixed_precision/marlin.py +139 -0
- vllm/model_executor/layers/quantization/kernels/scaled_mm/ScaledMMLinearKernel.py +67 -0
- vllm/model_executor/layers/quantization/kernels/scaled_mm/__init__.py +89 -0
- vllm/model_executor/layers/quantization/kernels/scaled_mm/aiter.py +161 -0
- vllm/model_executor/layers/quantization/kernels/scaled_mm/cpu.py +206 -0
- vllm/model_executor/layers/quantization/kernels/scaled_mm/cutlass.py +137 -0
- vllm/model_executor/layers/quantization/kernels/scaled_mm/triton.py +41 -0
- vllm/model_executor/layers/quantization/kernels/scaled_mm/xla.py +104 -0
- vllm/model_executor/layers/quantization/kv_cache.py +143 -0
- vllm/model_executor/layers/quantization/modelopt.py +1596 -0
- vllm/model_executor/layers/quantization/moe_wna16.py +484 -0
- vllm/model_executor/layers/quantization/mxfp4.py +988 -0
- vllm/model_executor/layers/quantization/petit.py +306 -0
- vllm/model_executor/layers/quantization/ptpc_fp8.py +129 -0
- vllm/model_executor/layers/quantization/quark/__init__.py +0 -0
- vllm/model_executor/layers/quantization/quark/quark.py +432 -0
- vllm/model_executor/layers/quantization/quark/quark_moe.py +561 -0
- vllm/model_executor/layers/quantization/quark/schemes/__init__.py +9 -0
- vllm/model_executor/layers/quantization/quark/schemes/quark_scheme.py +55 -0
- vllm/model_executor/layers/quantization/quark/schemes/quark_w4a4_mxfp4.py +239 -0
- vllm/model_executor/layers/quantization/quark/schemes/quark_w8a8_fp8.py +163 -0
- vllm/model_executor/layers/quantization/quark/schemes/quark_w8a8_int8.py +122 -0
- vllm/model_executor/layers/quantization/quark/utils.py +105 -0
- vllm/model_executor/layers/quantization/rtn.py +466 -0
- vllm/model_executor/layers/quantization/schema.py +86 -0
- vllm/model_executor/layers/quantization/torchao.py +214 -0
- vllm/model_executor/layers/quantization/tpu_int8.py +125 -0
- vllm/model_executor/layers/quantization/utils/__init__.py +6 -0
- vllm/model_executor/layers/quantization/utils/allspark_utils.py +52 -0
- vllm/model_executor/layers/quantization/utils/bitblas_utils.py +210 -0
- vllm/model_executor/layers/quantization/utils/configs/N=12288,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=12288,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=1536,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=1536,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=1536,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=1536,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=1536,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=1536,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=1536,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=1536,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=1536,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2048,K=512,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2048,K=512,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2048,K=512,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2048,K=512,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2048,K=512,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2048,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2048,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2048,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2048,K=512,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2112,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2112,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2304,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2304,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2304,K=7168,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2304,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2304,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2304,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2304,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2304,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=2304,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=1536,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=1536,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=24576,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=256,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=256,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=256,K=7168,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=256,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=256,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=256,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=256,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=256,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=1536,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=1536,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=1536,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=1536,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=1536,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=1536,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=1536,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=7168,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=3072,K=7168,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=32768,K=512,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=36864,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=36864,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=36864,K=7168,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=36864,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=36864,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=512,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=512,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=512,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=512,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=512,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=512,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4096,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4608,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4608,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4608,K=7168,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4608,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4608,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4608,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4608,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=4608,K=7168,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=512,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=512,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=512,K=7168,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=512,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=512,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=512,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=512,K=7168,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +18 -0
- vllm/model_executor/layers/quantization/utils/configs/N=576,K=7168,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1024,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1024,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1024,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1024,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1024,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1024,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1024,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1024,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1024,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1152,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1152,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1152,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1152,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1152,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1152,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1152,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1152,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=1152,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=128,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=128,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=128,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=128,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=128,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=128,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=16384,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=NVIDIA_A800-SXM4-80GB,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=18432,device_name=NVIDIA_L20Y,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2048,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2048,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2048,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2048,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2048,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2048,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2048,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2048,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2304,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2304,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2304,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2304,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2304,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2304,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2304,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=2304,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=256,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=256,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=256,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=256,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128,128].json +146 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=256,device_name=NVIDIA_L20,dtype=fp8_w8a8,block_shape=[128,128].json +26 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=8192,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=8192,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=7168,K=8192,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=8192,K=1536,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=8192,K=1536,device_name=AMD_Instinct_MI325X,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/N=8192,K=1536,device_name=AMD_Instinct_MI325_OAM,dtype=fp8_w8a8,block_shape=[128,128].json +164 -0
- vllm/model_executor/layers/quantization/utils/configs/README.md +3 -0
- vllm/model_executor/layers/quantization/utils/flashinfer_fp4_moe.py +79 -0
- vllm/model_executor/layers/quantization/utils/flashinfer_utils.py +248 -0
- vllm/model_executor/layers/quantization/utils/fp8_utils.py +949 -0
- vllm/model_executor/layers/quantization/utils/gptq_utils.py +146 -0
- vllm/model_executor/layers/quantization/utils/int8_utils.py +492 -0
- vllm/model_executor/layers/quantization/utils/layer_utils.py +40 -0
- vllm/model_executor/layers/quantization/utils/machete_utils.py +50 -0
- vllm/model_executor/layers/quantization/utils/marlin_utils.py +479 -0
- vllm/model_executor/layers/quantization/utils/marlin_utils_fp4.py +396 -0
- vllm/model_executor/layers/quantization/utils/marlin_utils_fp8.py +345 -0
- vllm/model_executor/layers/quantization/utils/marlin_utils_test.py +165 -0
- vllm/model_executor/layers/quantization/utils/marlin_utils_test_24.py +464 -0
- vllm/model_executor/layers/quantization/utils/mxfp4_utils.py +141 -0
- vllm/model_executor/layers/quantization/utils/mxfp8_utils.py +20 -0
- vllm/model_executor/layers/quantization/utils/nvfp4_emulation_utils.py +137 -0
- vllm/model_executor/layers/quantization/utils/nvfp4_moe_support.py +59 -0
- vllm/model_executor/layers/quantization/utils/petit_utils.py +122 -0
- vllm/model_executor/layers/quantization/utils/quant_utils.py +641 -0
- vllm/model_executor/layers/quantization/utils/w8a8_utils.py +458 -0
- vllm/model_executor/layers/resampler.py +270 -0
- vllm/model_executor/layers/rotary_embedding/__init__.py +204 -0
- vllm/model_executor/layers/rotary_embedding/base.py +177 -0
- vllm/model_executor/layers/rotary_embedding/common.py +150 -0
- vllm/model_executor/layers/rotary_embedding/deepseek_scaling_rope.py +138 -0
- vllm/model_executor/layers/rotary_embedding/dual_chunk_rope.py +197 -0
- vllm/model_executor/layers/rotary_embedding/dynamic_ntk_alpha_rope.py +41 -0
- vllm/model_executor/layers/rotary_embedding/dynamic_ntk_scaling_rope.py +67 -0
- vllm/model_executor/layers/rotary_embedding/ernie45_vl_rope.py +80 -0
- vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py +115 -0
- vllm/model_executor/layers/rotary_embedding/llama3_rope.py +54 -0
- vllm/model_executor/layers/rotary_embedding/llama4_vision_rope.py +81 -0
- vllm/model_executor/layers/rotary_embedding/mrope.py +1321 -0
- vllm/model_executor/layers/rotary_embedding/ntk_scaling_rope.py +42 -0
- vllm/model_executor/layers/rotary_embedding/phi3_long_rope_scaled_rope.py +129 -0
- vllm/model_executor/layers/rotary_embedding/rocm_aiter_rope_ops.py +86 -0
- vllm/model_executor/layers/rotary_embedding/yarn_scaling_rope.py +68 -0
- vllm/model_executor/layers/shared_fused_moe/__init__.py +6 -0
- vllm/model_executor/layers/shared_fused_moe/shared_fused_moe.py +56 -0
- vllm/model_executor/layers/utils.py +195 -0
- vllm/model_executor/layers/vocab_parallel_embedding.py +487 -0
- vllm/model_executor/model_loader/__init__.py +138 -0
- vllm/model_executor/model_loader/base_loader.py +52 -0
- vllm/model_executor/model_loader/bitsandbytes_loader.py +788 -0
- vllm/model_executor/model_loader/default_loader.py +277 -0
- vllm/model_executor/model_loader/dummy_loader.py +28 -0
- vllm/model_executor/model_loader/gguf_loader.py +155 -0
- vllm/model_executor/model_loader/runai_streamer_loader.py +104 -0
- vllm/model_executor/model_loader/sharded_state_loader.py +199 -0
- vllm/model_executor/model_loader/tensorizer.py +738 -0
- vllm/model_executor/model_loader/tensorizer_loader.py +143 -0
- vllm/model_executor/model_loader/tpu.py +114 -0
- vllm/model_executor/model_loader/utils.py +292 -0
- vllm/model_executor/model_loader/weight_utils.py +990 -0
- vllm/model_executor/models/__init__.py +33 -0
- vllm/model_executor/models/adapters.py +542 -0
- vllm/model_executor/models/aimv2.py +246 -0
- vllm/model_executor/models/apertus.py +579 -0
- vllm/model_executor/models/arcee.py +422 -0
- vllm/model_executor/models/arctic.py +558 -0
- vllm/model_executor/models/aria.py +650 -0
- vllm/model_executor/models/aya_vision.py +468 -0
- vllm/model_executor/models/baichuan.py +474 -0
- vllm/model_executor/models/bailing_moe.py +642 -0
- vllm/model_executor/models/bamba.py +514 -0
- vllm/model_executor/models/bert.py +665 -0
- vllm/model_executor/models/bert_with_rope.py +687 -0
- vllm/model_executor/models/blip.py +339 -0
- vllm/model_executor/models/blip2.py +712 -0
- vllm/model_executor/models/bloom.py +374 -0
- vllm/model_executor/models/chameleon.py +1139 -0
- vllm/model_executor/models/chatglm.py +476 -0
- vllm/model_executor/models/clip.py +407 -0
- vllm/model_executor/models/cohere2_vision.py +481 -0
- vllm/model_executor/models/commandr.py +465 -0
- vllm/model_executor/models/config.py +445 -0
- vllm/model_executor/models/dbrx.py +471 -0
- vllm/model_executor/models/deepseek.py +497 -0
- vllm/model_executor/models/deepseek_eagle.py +240 -0
- vllm/model_executor/models/deepseek_mtp.py +289 -0
- vllm/model_executor/models/deepseek_v2.py +1444 -0
- vllm/model_executor/models/deepseek_vl2.py +658 -0
- vllm/model_executor/models/dots1.py +546 -0
- vllm/model_executor/models/dots_ocr.py +873 -0
- vllm/model_executor/models/ernie45.py +43 -0
- vllm/model_executor/models/ernie45_moe.py +607 -0
- vllm/model_executor/models/ernie45_vl.py +1527 -0
- vllm/model_executor/models/ernie45_vl_moe.py +727 -0
- vllm/model_executor/models/ernie_mtp.py +268 -0
- vllm/model_executor/models/exaone.py +550 -0
- vllm/model_executor/models/exaone4.py +533 -0
- vllm/model_executor/models/fairseq2_llama.py +154 -0
- vllm/model_executor/models/falcon.py +509 -0
- vllm/model_executor/models/falcon_h1.py +674 -0
- vllm/model_executor/models/fuyu.py +399 -0
- vllm/model_executor/models/gemma.py +425 -0
- vllm/model_executor/models/gemma2.py +422 -0
- vllm/model_executor/models/gemma3.py +555 -0
- vllm/model_executor/models/gemma3_mm.py +721 -0
- vllm/model_executor/models/gemma3n.py +1113 -0
- vllm/model_executor/models/gemma3n_mm.py +761 -0
- vllm/model_executor/models/glm.py +23 -0
- vllm/model_executor/models/glm4.py +304 -0
- vllm/model_executor/models/glm4_1v.py +1690 -0
- vllm/model_executor/models/glm4_moe.py +727 -0
- vllm/model_executor/models/glm4_moe_mtp.py +301 -0
- vllm/model_executor/models/glm4v.py +654 -0
- vllm/model_executor/models/gpt2.py +380 -0
- vllm/model_executor/models/gpt_bigcode.py +344 -0
- vllm/model_executor/models/gpt_j.py +339 -0
- vllm/model_executor/models/gpt_neox.py +330 -0
- vllm/model_executor/models/gpt_oss.py +712 -0
- vllm/model_executor/models/granite.py +489 -0
- vllm/model_executor/models/granite_speech.py +794 -0
- vllm/model_executor/models/granitemoe.py +550 -0
- vllm/model_executor/models/granitemoehybrid.py +614 -0
- vllm/model_executor/models/granitemoeshared.py +332 -0
- vllm/model_executor/models/gritlm.py +262 -0
- vllm/model_executor/models/grok1.py +547 -0
- vllm/model_executor/models/h2ovl.py +536 -0
- vllm/model_executor/models/hunyuan_v1.py +1042 -0
- vllm/model_executor/models/hyperclovax_vision.py +1192 -0
- vllm/model_executor/models/idefics2_vision_model.py +417 -0
- vllm/model_executor/models/idefics3.py +756 -0
- vllm/model_executor/models/interfaces.py +959 -0
- vllm/model_executor/models/interfaces_base.py +192 -0
- vllm/model_executor/models/intern_vit.py +441 -0
- vllm/model_executor/models/internlm2.py +450 -0
- vllm/model_executor/models/internlm2_ve.py +148 -0
- vllm/model_executor/models/interns1.py +838 -0
- vllm/model_executor/models/interns1_vit.py +418 -0
- vllm/model_executor/models/internvl.py +1423 -0
- vllm/model_executor/models/jais.py +373 -0
- vllm/model_executor/models/jamba.py +591 -0
- vllm/model_executor/models/jina_vl.py +144 -0
- vllm/model_executor/models/keye.py +1680 -0
- vllm/model_executor/models/keye_vl1_5.py +602 -0
- vllm/model_executor/models/kimi_vl.py +618 -0
- vllm/model_executor/models/lfm2.py +548 -0
- vllm/model_executor/models/llama.py +669 -0
- vllm/model_executor/models/llama4.py +746 -0
- vllm/model_executor/models/llama4_eagle.py +239 -0
- vllm/model_executor/models/llama_eagle.py +179 -0
- vllm/model_executor/models/llama_eagle3.py +296 -0
- vllm/model_executor/models/llava.py +870 -0
- vllm/model_executor/models/llava_next.py +571 -0
- vllm/model_executor/models/llava_next_video.py +476 -0
- vllm/model_executor/models/llava_onevision.py +942 -0
- vllm/model_executor/models/longcat_flash.py +715 -0
- vllm/model_executor/models/longcat_flash_mtp.py +352 -0
- vllm/model_executor/models/mamba.py +275 -0
- vllm/model_executor/models/mamba2.py +291 -0
- vllm/model_executor/models/medusa.py +169 -0
- vllm/model_executor/models/midashenglm.py +792 -0
- vllm/model_executor/models/mimo.py +188 -0
- vllm/model_executor/models/mimo_mtp.py +280 -0
- vllm/model_executor/models/minicpm.py +631 -0
- vllm/model_executor/models/minicpm3.py +230 -0
- vllm/model_executor/models/minicpm_eagle.py +389 -0
- vllm/model_executor/models/minicpmo.py +770 -0
- vllm/model_executor/models/minicpmv.py +1784 -0
- vllm/model_executor/models/minimax_text_01.py +986 -0
- vllm/model_executor/models/minimax_vl_01.py +426 -0
- vllm/model_executor/models/mistral3.py +628 -0
- vllm/model_executor/models/mixtral.py +606 -0
- vllm/model_executor/models/mllama4.py +1076 -0
- vllm/model_executor/models/mlp_speculator.py +206 -0
- vllm/model_executor/models/modernbert.py +374 -0
- vllm/model_executor/models/module_mapping.py +72 -0
- vllm/model_executor/models/molmo.py +1567 -0
- vllm/model_executor/models/moonvit.py +673 -0
- vllm/model_executor/models/motif.py +345 -0
- vllm/model_executor/models/mpt.py +329 -0
- vllm/model_executor/models/nano_nemotron_vl.py +1394 -0
- vllm/model_executor/models/nemotron.py +507 -0
- vllm/model_executor/models/nemotron_h.py +565 -0
- vllm/model_executor/models/nemotron_nas.py +481 -0
- vllm/model_executor/models/nemotron_vl.py +652 -0
- vllm/model_executor/models/nvlm_d.py +203 -0
- vllm/model_executor/models/olmo.py +404 -0
- vllm/model_executor/models/olmo2.py +439 -0
- vllm/model_executor/models/olmoe.py +483 -0
- vllm/model_executor/models/opt.py +412 -0
- vllm/model_executor/models/orion.py +348 -0
- vllm/model_executor/models/ovis.py +559 -0
- vllm/model_executor/models/ovis2_5.py +642 -0
- vllm/model_executor/models/paligemma.py +411 -0
- vllm/model_executor/models/persimmon.py +343 -0
- vllm/model_executor/models/phi.py +356 -0
- vllm/model_executor/models/phi3.py +19 -0
- vllm/model_executor/models/phi3v.py +698 -0
- vllm/model_executor/models/phi4_multimodal.py +1475 -0
- vllm/model_executor/models/phi4mm.py +1279 -0
- vllm/model_executor/models/phi4mm_audio.py +1254 -0
- vllm/model_executor/models/phi4mm_utils.py +1875 -0
- vllm/model_executor/models/phimoe.py +679 -0
- vllm/model_executor/models/pixtral.py +1345 -0
- vllm/model_executor/models/plamo2.py +978 -0
- vllm/model_executor/models/qwen.py +361 -0
- vllm/model_executor/models/qwen2.py +523 -0
- vllm/model_executor/models/qwen2_5_omni_thinker.py +984 -0
- vllm/model_executor/models/qwen2_5_vl.py +1481 -0
- vllm/model_executor/models/qwen2_audio.py +489 -0
- vllm/model_executor/models/qwen2_moe.py +558 -0
- vllm/model_executor/models/qwen2_rm.py +122 -0
- vllm/model_executor/models/qwen2_vl.py +1670 -0
- vllm/model_executor/models/qwen3.py +341 -0
- vllm/model_executor/models/qwen3_moe.py +692 -0
- vllm/model_executor/models/qwen3_next.py +1266 -0
- vllm/model_executor/models/qwen3_next_mtp.py +281 -0
- vllm/model_executor/models/qwen3_vl.py +1613 -0
- vllm/model_executor/models/qwen3_vl_moe.py +358 -0
- vllm/model_executor/models/qwen_vl.py +795 -0
- vllm/model_executor/models/radio.py +576 -0
- vllm/model_executor/models/registry.py +990 -0
- vllm/model_executor/models/roberta.py +252 -0
- vllm/model_executor/models/rvl.py +103 -0
- vllm/model_executor/models/seed_oss.py +485 -0
- vllm/model_executor/models/siglip.py +540 -0
- vllm/model_executor/models/siglip2navit.py +689 -0
- vllm/model_executor/models/skyworkr1v.py +911 -0
- vllm/model_executor/models/smolvlm.py +44 -0
- vllm/model_executor/models/solar.py +504 -0
- vllm/model_executor/models/stablelm.py +341 -0
- vllm/model_executor/models/starcoder2.py +354 -0
- vllm/model_executor/models/step3_text.py +510 -0
- vllm/model_executor/models/step3_vl.py +1072 -0
- vllm/model_executor/models/swin.py +475 -0
- vllm/model_executor/models/tarsier.py +639 -0
- vllm/model_executor/models/telechat2.py +151 -0
- vllm/model_executor/models/teleflm.py +79 -0
- vllm/model_executor/models/terratorch.py +294 -0
- vllm/model_executor/models/transformers.py +948 -0
- vllm/model_executor/models/ultravox.py +654 -0
- vllm/model_executor/models/utils.py +808 -0
- vllm/model_executor/models/vision.py +404 -0
- vllm/model_executor/models/voxtral.py +786 -0
- vllm/model_executor/models/whisper.py +963 -0
- vllm/model_executor/models/zamba2.py +960 -0
- vllm/model_executor/parameter.py +620 -0
- vllm/model_executor/utils.py +86 -0
- vllm/model_executor/warmup/__init__.py +0 -0
- vllm/model_executor/warmup/deep_gemm_warmup.py +230 -0
- vllm/model_executor/warmup/kernel_warmup.py +83 -0
- vllm/multimodal/__init__.py +33 -0
- vllm/multimodal/audio.py +116 -0
- vllm/multimodal/base.py +27 -0
- vllm/multimodal/cache.py +697 -0
- vllm/multimodal/evs.py +273 -0
- vllm/multimodal/hasher.py +102 -0
- vllm/multimodal/image.py +130 -0
- vllm/multimodal/inputs.py +987 -0
- vllm/multimodal/parse.py +511 -0
- vllm/multimodal/processing.py +2148 -0
- vllm/multimodal/profiling.py +284 -0
- vllm/multimodal/registry.py +345 -0
- vllm/multimodal/utils.py +503 -0
- vllm/multimodal/video.py +319 -0
- vllm/outputs.py +324 -0
- vllm/platforms/__init__.py +263 -0
- vllm/platforms/cpu.py +340 -0
- vllm/platforms/cuda.py +668 -0
- vllm/platforms/interface.py +620 -0
- vllm/platforms/rocm.py +497 -0
- vllm/platforms/tpu.py +233 -0
- vllm/platforms/xpu.py +243 -0
- vllm/plugins/__init__.py +72 -0
- vllm/plugins/io_processors/__init__.py +68 -0
- vllm/plugins/io_processors/interface.py +67 -0
- vllm/plugins/lora_resolvers/README.md +16 -0
- vllm/plugins/lora_resolvers/__init__.py +0 -0
- vllm/plugins/lora_resolvers/filesystem_resolver.py +50 -0
- vllm/pooling_params.py +191 -0
- vllm/profiler/__init__.py +0 -0
- vllm/profiler/layerwise_profile.py +375 -0
- vllm/profiler/utils.py +148 -0
- vllm/py.typed +2 -0
- vllm/ray/__init__.py +0 -0
- vllm/ray/lazy_utils.py +22 -0
- vllm/ray/ray_env.py +72 -0
- vllm/reasoning/__init__.py +29 -0
- vllm/reasoning/abs_reasoning_parsers.py +202 -0
- vllm/reasoning/basic_parsers.py +156 -0
- vllm/reasoning/deepseek_r1_reasoning_parser.py +67 -0
- vllm/reasoning/glm4_moe_reasoning_parser.py +151 -0
- vllm/reasoning/gptoss_reasoning_parser.py +87 -0
- vllm/reasoning/granite_reasoning_parser.py +363 -0
- vllm/reasoning/hunyuan_a13b_reasoning_parser.py +245 -0
- vllm/reasoning/mistral_reasoning_parser.py +56 -0
- vllm/reasoning/qwen3_reasoning_parser.py +72 -0
- vllm/reasoning/seedoss_reasoning_parser.py +28 -0
- vllm/reasoning/step3_reasoning_parser.py +109 -0
- vllm/sampling_params.py +593 -0
- vllm/scalar_type.py +349 -0
- vllm/scripts.py +15 -0
- vllm/sequence.py +103 -0
- vllm/tasks.py +11 -0
- vllm/test_utils.py +129 -0
- vllm/third_party/__init__.py +0 -0
- vllm/third_party/pynvml.py +6140 -0
- vllm/tracing.py +136 -0
- vllm/transformers_utils/__init__.py +24 -0
- vllm/transformers_utils/chat_templates/__init__.py +5 -0
- vllm/transformers_utils/chat_templates/registry.py +70 -0
- vllm/transformers_utils/chat_templates/template_basic.jinja +3 -0
- vllm/transformers_utils/chat_templates/template_blip2.jinja +11 -0
- vllm/transformers_utils/chat_templates/template_chatml.jinja +10 -0
- vllm/transformers_utils/chat_templates/template_deepseek_vl2.jinja +23 -0
- vllm/transformers_utils/chat_templates/template_fuyu.jinja +3 -0
- vllm/transformers_utils/chat_templates/template_minicpmv45.jinja +93 -0
- vllm/transformers_utils/config.py +1102 -0
- vllm/transformers_utils/config_parser_base.py +20 -0
- vllm/transformers_utils/configs/__init__.py +63 -0
- vllm/transformers_utils/configs/arctic.py +207 -0
- vllm/transformers_utils/configs/chatglm.py +72 -0
- vllm/transformers_utils/configs/deepseek_v3.py +101 -0
- vllm/transformers_utils/configs/deepseek_vl2.py +216 -0
- vllm/transformers_utils/configs/dotsocr.py +69 -0
- vllm/transformers_utils/configs/eagle.py +84 -0
- vllm/transformers_utils/configs/falcon.py +90 -0
- vllm/transformers_utils/configs/jais.py +237 -0
- vllm/transformers_utils/configs/kimi_vl.py +37 -0
- vllm/transformers_utils/configs/medusa.py +63 -0
- vllm/transformers_utils/configs/midashenglm.py +101 -0
- vllm/transformers_utils/configs/mistral.py +165 -0
- vllm/transformers_utils/configs/mlp_speculator.py +68 -0
- vllm/transformers_utils/configs/moonvit.py +33 -0
- vllm/transformers_utils/configs/nemotron.py +205 -0
- vllm/transformers_utils/configs/nemotron_h.py +259 -0
- vllm/transformers_utils/configs/nemotron_vl.py +56 -0
- vllm/transformers_utils/configs/olmo3.py +80 -0
- vllm/transformers_utils/configs/ovis.py +176 -0
- vllm/transformers_utils/configs/qwen3_next.py +275 -0
- vllm/transformers_utils/configs/radio.py +91 -0
- vllm/transformers_utils/configs/speculators/__init__.py +2 -0
- vllm/transformers_utils/configs/speculators/algos.py +32 -0
- vllm/transformers_utils/configs/speculators/base.py +111 -0
- vllm/transformers_utils/configs/step3_vl.py +123 -0
- vllm/transformers_utils/configs/ultravox.py +116 -0
- vllm/transformers_utils/detokenizer_utils.py +199 -0
- vllm/transformers_utils/dynamic_module.py +60 -0
- vllm/transformers_utils/processor.py +299 -0
- vllm/transformers_utils/processors/__init__.py +16 -0
- vllm/transformers_utils/processors/deepseek_vl2.py +362 -0
- vllm/transformers_utils/processors/ovis.py +420 -0
- vllm/transformers_utils/processors/ovis2_5.py +458 -0
- vllm/transformers_utils/runai_utils.py +104 -0
- vllm/transformers_utils/s3_utils.py +93 -0
- vllm/transformers_utils/tokenizer.py +292 -0
- vllm/transformers_utils/tokenizer_base.py +154 -0
- vllm/transformers_utils/tokenizers/__init__.py +10 -0
- vllm/transformers_utils/tokenizers/mistral.py +521 -0
- vllm/transformers_utils/utils.py +108 -0
- vllm/triton_utils/__init__.py +16 -0
- vllm/triton_utils/importing.py +96 -0
- vllm/usage/__init__.py +0 -0
- vllm/usage/usage_lib.py +259 -0
- vllm/utils/__init__.py +3566 -0
- vllm/utils/deep_gemm.py +319 -0
- vllm/utils/flashinfer.py +443 -0
- vllm/utils/jsontree.py +178 -0
- vllm/utils/tensor_schema.py +235 -0
- vllm/v1/__init__.py +0 -0
- vllm/v1/attention/__init__.py +0 -0
- vllm/v1/attention/backends/__init__.py +0 -0
- vllm/v1/attention/backends/cpu_attn.py +919 -0
- vllm/v1/attention/backends/flash_attn.py +795 -0
- vllm/v1/attention/backends/flashinfer.py +1181 -0
- vllm/v1/attention/backends/flex_attention.py +861 -0
- vllm/v1/attention/backends/gdn_attn.py +332 -0
- vllm/v1/attention/backends/linear_attn.py +67 -0
- vllm/v1/attention/backends/mamba1_attn.py +81 -0
- vllm/v1/attention/backends/mamba2_attn.py +232 -0
- vllm/v1/attention/backends/mamba_attn.py +52 -0
- vllm/v1/attention/backends/mla/__init__.py +0 -0
- vllm/v1/attention/backends/mla/common.py +1783 -0
- vllm/v1/attention/backends/mla/cutlass_mla.py +248 -0
- vllm/v1/attention/backends/mla/flashattn_mla.py +271 -0
- vllm/v1/attention/backends/mla/flashinfer_mla.py +114 -0
- vllm/v1/attention/backends/mla/flashmla.py +203 -0
- vllm/v1/attention/backends/mla/flashmla_sparse.py +544 -0
- vllm/v1/attention/backends/mla/indexer.py +342 -0
- vllm/v1/attention/backends/mla/rocm_aiter_mla.py +255 -0
- vllm/v1/attention/backends/mla/triton_mla.py +177 -0
- vllm/v1/attention/backends/pallas.py +409 -0
- vllm/v1/attention/backends/rocm_aiter_fa.py +549 -0
- vllm/v1/attention/backends/rocm_attn.py +426 -0
- vllm/v1/attention/backends/short_conv_attn.py +94 -0
- vllm/v1/attention/backends/tree_attn.py +451 -0
- vllm/v1/attention/backends/triton_attn.py +361 -0
- vllm/v1/attention/backends/utils.py +990 -0
- vllm/v1/attention/backends/xformers.py +438 -0
- vllm/v1/core/__init__.py +0 -0
- vllm/v1/core/block_pool.py +416 -0
- vllm/v1/core/encoder_cache_manager.py +333 -0
- vllm/v1/core/kv_cache_coordinator.py +440 -0
- vllm/v1/core/kv_cache_manager.py +399 -0
- vllm/v1/core/kv_cache_utils.py +1291 -0
- vllm/v1/core/sched/__init__.py +0 -0
- vllm/v1/core/sched/async_scheduler.py +47 -0
- vllm/v1/core/sched/interface.py +158 -0
- vllm/v1/core/sched/output.py +166 -0
- vllm/v1/core/sched/request_queue.py +224 -0
- vllm/v1/core/sched/scheduler.py +1296 -0
- vllm/v1/core/sched/utils.py +69 -0
- vllm/v1/core/single_type_kv_cache_manager.py +671 -0
- vllm/v1/cudagraph_dispatcher.py +125 -0
- vllm/v1/engine/__init__.py +203 -0
- vllm/v1/engine/async_llm.py +742 -0
- vllm/v1/engine/coordinator.py +357 -0
- vllm/v1/engine/core.py +1235 -0
- vllm/v1/engine/core_client.py +1334 -0
- vllm/v1/engine/detokenizer.py +349 -0
- vllm/v1/engine/exceptions.py +17 -0
- vllm/v1/engine/llm_engine.py +370 -0
- vllm/v1/engine/logprobs.py +201 -0
- vllm/v1/engine/output_processor.py +576 -0
- vllm/v1/engine/parallel_sampling.py +133 -0
- vllm/v1/engine/processor.py +545 -0
- vllm/v1/engine/utils.py +860 -0
- vllm/v1/executor/__init__.py +0 -0
- vllm/v1/executor/abstract.py +137 -0
- vllm/v1/executor/multiproc_executor.py +726 -0
- vllm/v1/executor/ray_distributed_executor.py +108 -0
- vllm/v1/executor/utils.py +23 -0
- vllm/v1/kv_cache_interface.py +375 -0
- vllm/v1/kv_offload/__init__.py +0 -0
- vllm/v1/kv_offload/abstract.py +165 -0
- vllm/v1/kv_offload/backend.py +96 -0
- vllm/v1/kv_offload/backends/__init__.py +0 -0
- vllm/v1/kv_offload/backends/cpu.py +61 -0
- vllm/v1/kv_offload/cpu.py +75 -0
- vllm/v1/kv_offload/factory.py +56 -0
- vllm/v1/kv_offload/lru_manager.py +132 -0
- vllm/v1/kv_offload/mediums.py +39 -0
- vllm/v1/kv_offload/spec.py +61 -0
- vllm/v1/kv_offload/worker/__init__.py +0 -0
- vllm/v1/kv_offload/worker/cpu_gpu.py +171 -0
- vllm/v1/kv_offload/worker/worker.py +142 -0
- vllm/v1/metrics/__init__.py +0 -0
- vllm/v1/metrics/loggers.py +741 -0
- vllm/v1/metrics/prometheus.py +82 -0
- vllm/v1/metrics/ray_wrappers.py +152 -0
- vllm/v1/metrics/reader.py +246 -0
- vllm/v1/metrics/stats.py +257 -0
- vllm/v1/outputs.py +161 -0
- vllm/v1/pool/__init__.py +0 -0
- vllm/v1/pool/metadata.py +77 -0
- vllm/v1/request.py +241 -0
- vllm/v1/sample/__init__.py +0 -0
- vllm/v1/sample/logits_processor/__init__.py +294 -0
- vllm/v1/sample/logits_processor/builtin.py +275 -0
- vllm/v1/sample/logits_processor/interface.py +97 -0
- vllm/v1/sample/logits_processor/state.py +161 -0
- vllm/v1/sample/metadata.py +43 -0
- vllm/v1/sample/ops/__init__.py +0 -0
- vllm/v1/sample/ops/bad_words.py +39 -0
- vllm/v1/sample/ops/logprobs.py +26 -0
- vllm/v1/sample/ops/penalties.py +43 -0
- vllm/v1/sample/ops/topk_topp_sampler.py +292 -0
- vllm/v1/sample/rejection_sampler.py +623 -0
- vllm/v1/sample/sampler.py +285 -0
- vllm/v1/sample/tpu/__init__.py +0 -0
- vllm/v1/sample/tpu/metadata.py +124 -0
- vllm/v1/sample/tpu/sampler.py +213 -0
- vllm/v1/serial_utils.py +423 -0
- vllm/v1/spec_decode/__init__.py +0 -0
- vllm/v1/spec_decode/eagle.py +1011 -0
- vllm/v1/spec_decode/medusa.py +66 -0
- vllm/v1/spec_decode/metadata.py +62 -0
- vllm/v1/spec_decode/metrics.py +211 -0
- vllm/v1/spec_decode/ngram_proposer.py +276 -0
- vllm/v1/spec_decode/utils.py +14 -0
- vllm/v1/structured_output/__init__.py +295 -0
- vllm/v1/structured_output/backend_guidance.py +245 -0
- vllm/v1/structured_output/backend_lm_format_enforcer.py +167 -0
- vllm/v1/structured_output/backend_outlines.py +320 -0
- vllm/v1/structured_output/backend_types.py +134 -0
- vllm/v1/structured_output/backend_xgrammar.py +327 -0
- vllm/v1/structured_output/request.py +86 -0
- vllm/v1/structured_output/utils.py +454 -0
- vllm/v1/utils.py +396 -0
- vllm/v1/worker/__init__.py +0 -0
- vllm/v1/worker/block_table.py +210 -0
- vllm/v1/worker/cpu_model_runner.py +175 -0
- vllm/v1/worker/cpu_worker.py +156 -0
- vllm/v1/worker/gpu_input_batch.py +863 -0
- vllm/v1/worker/gpu_model_runner.py +4160 -0
- vllm/v1/worker/gpu_ubatch_wrapper.py +399 -0
- vllm/v1/worker/gpu_worker.py +710 -0
- vllm/v1/worker/kv_connector_model_runner_mixin.py +132 -0
- vllm/v1/worker/lora_model_runner_mixin.py +183 -0
- vllm/v1/worker/tpu_input_batch.py +587 -0
- vllm/v1/worker/tpu_model_runner.py +1946 -0
- vllm/v1/worker/tpu_worker.py +346 -0
- vllm/v1/worker/ubatch_splitting.py +192 -0
- vllm/v1/worker/ubatch_utils.py +27 -0
- vllm/v1/worker/ubatching.py +224 -0
- vllm/v1/worker/utils.py +344 -0
- vllm/v1/worker/worker_base.py +65 -0
- vllm/v1/worker/xpu_model_runner.py +57 -0
- vllm/v1/worker/xpu_worker.py +179 -0
- vllm/version.py +41 -0
- vllm/vllm_flash_attn/.gitkeep +0 -0
- vllm/worker/__init__.py +0 -0
- vllm/worker/worker_base.py +279 -0
- vllm_cpu-0.11.0.post2.dist-info/METADATA +348 -0
- vllm_cpu-0.11.0.post2.dist-info/RECORD +1398 -0
- vllm_cpu-0.11.0.post2.dist-info/WHEEL +5 -0
- vllm_cpu-0.11.0.post2.dist-info/entry_points.txt +5 -0
- vllm_cpu-0.11.0.post2.dist-info/top_level.txt +1 -0
|
@@ -0,0 +1,19 @@
|
|
|
1
|
+
# SPDX-License-Identifier: Apache-2.0
|
|
2
|
+
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
|
3
|
+
|
|
4
|
+
# Adapted from llama.py
|
|
5
|
+
"""Inference-only Phi3 model code inherit from Llama.py"""
|
|
6
|
+
|
|
7
|
+
from vllm.model_executor.models.llama import LlamaForCausalLM
|
|
8
|
+
|
|
9
|
+
|
|
10
|
+
class Phi3ForCausalLM(LlamaForCausalLM):
|
|
11
|
+
|
|
12
|
+
packed_modules_mapping = {
|
|
13
|
+
"qkv_proj": [
|
|
14
|
+
"qkv_proj",
|
|
15
|
+
],
|
|
16
|
+
"gate_up_proj": [
|
|
17
|
+
"gate_up_proj",
|
|
18
|
+
],
|
|
19
|
+
}
|