TensorRT-LLMs/cpp/tensorrt_llm/kernels/trtllmGenKernels/blockScaleMoe
Nikita Korobov f2ebaf288a
[None][feat] TRT-LLM Gen MoE optimize DeepSeek Fp8 activation kernel (#9175)
Signed-off-by: Nikita Korobov <14355239+nekorobov@users.noreply.github.com>
2025-11-21 15:35:00 +01:00
..
CMakeLists.txt feat: [Deepseek] Add trtllm-gen MOE FP4 MOE backend (#3387) 2025-04-21 10:01:33 +08:00
DevKernel.cu [None][feat] TRT-LLM Gen MoE optimize DeepSeek Fp8 activation kernel (#9175) 2025-11-21 15:35:00 +01:00
DevKernel.h [None][feat] TRT-LLM Gen MoE optimize DeepSeek Fp8 activation kernel (#9175) 2025-11-21 15:35:00 +01:00
IntFastDiv.h [fix] Fix comment to pass guardwords check (#5191) 2025-06-13 15:49:59 +08:00
RoutingDeepSeek.cu [None][fix] Update the default invalid value for deepseek mode of routing (#9222) 2025-11-19 10:14:06 +08:00
RoutingKernel.cuh [TRTLLM-9286][feat] Integration of CuteDSL NVFP4 grouped GEMM (#8880) 2025-11-18 17:40:12 -08:00
RoutingKernel.h [TRTLLM-9286][feat] Integration of CuteDSL NVFP4 grouped GEMM (#8880) 2025-11-18 17:40:12 -08:00
RoutingKernelTopK.cuh [TRTLLM-8637][feat] Optimize the routing kernel for DeepseekV3 (MoE CUTLASS backend); Add support for KimiK2 and Qwen-next (MoE TRTLLM backend) (#7761) 2025-10-20 10:08:31 +08:00
RoutingLlama4.cu [TRTLLM-9286][feat] Integration of CuteDSL NVFP4 grouped GEMM (#8880) 2025-11-18 17:40:12 -08:00
RoutingRenormalize.cu [TRTLLM-9286][feat] Integration of CuteDSL NVFP4 grouped GEMM (#8880) 2025-11-18 17:40:12 -08:00
runner.cu [TRTLLM-9286][feat] Integration of CuteDSL NVFP4 grouped GEMM (#8880) 2025-11-18 17:40:12 -08:00
runner.h [None][feat] Update TRTLLM MoE cubins; reduce mxfp4 weight padding requirement; tighten TMA bound (#9025) 2025-11-17 10:04:29 +08:00