vllm/csrc
Divakar Verma a66cf40b20
[Kernel][ROCm][AMD] enable fused topk_softmax kernel for moe layer (#4927)
This PR enables the fused topk_softmax kernel used in moe layer for HIP
2024-06-02 14:13:26 -07:00
..
attention [Model] Support MAP-NEO model (#5081) 2024-05-30 19:24:41 -07:00
cpu [Model] Support MAP-NEO model (#5081) 2024-05-30 19:24:41 -07:00
moe [Kernel][ROCm][AMD] enable fused topk_softmax kernel for moe layer (#4927) 2024-06-02 14:13:26 -07:00
punica [Model] LoRA gptbigcode implementation (#3949) 2024-05-22 13:58:59 -07:00
quantization [Kernel] Update Cutlass fp8 configs (#5144) 2024-06-01 08:46:07 +00:00
activation_kernels.cu [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
cache_kernels.cu [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
cache.h [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
cuda_compat.h [Kernel][ROCm][AMD] enable fused topk_softmax kernel for moe layer (#4927) 2024-06-02 14:13:26 -07:00
cuda_utils_kernels.cu [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
cuda_utils.h [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
custom_all_reduce_test.cu [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
custom_all_reduce.cu [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
custom_all_reduce.cuh [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
dispatch_utils.h [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
layernorm_kernels.cu [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
moe_align_block_size_kernels.cu [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
ops.h [Kernel][Backend][Model] Blocksparse flash attention kernel and Phi-3-Small model (#4799) 2024-05-24 22:00:52 -07:00
pos_encoding_kernels.cu [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00
pybind.cpp [Kernel] Initial Activation Quantization Support (#4525) 2024-05-23 21:29:18 +00:00
reduction_utils.cuh [CI/Build] Enforce style for C++ and CUDA code with clang-format (#4722) 2024-05-22 07:18:41 +00:00