| .. |
|
attention
|
[NVIDIA] Add Cutlass MLA backend (#17625)
|
2025-06-03 21:40:26 -07:00 |
|
core
|
[Kernel] fp4 marlin kernel (#17687)
|
2025-05-10 19:58:49 -07:00 |
|
cpu
|
[Bugfix] Add half type support in reshape_and_cache_cpu_impl on x86 cpu platform (#18430)
|
2025-05-23 01:41:37 -07:00 |
|
cutlass_extensions
|
[Misc] Add SPDX-FileCopyrightText (#19100)
|
2025-06-03 11:20:17 -07:00 |
|
mamba
|
feat(rocm-support): support mamba2 on rocm (#18565)
|
2025-05-27 00:07:53 -07:00 |
|
moe
|
[Kernel] DeepEP dispatch-combine kernel integration (#18434)
|
2025-06-03 12:30:02 -07:00 |
|
prepare_inputs
|
[Misc][Easy] Annotate unused vars in the csrc files (#14798)
|
2025-03-15 12:40:09 +08:00 |
|
quantization
|
[Perf] Tune `scaled_fp8_quant` by increasing vectorization (#18844)
|
2025-06-03 13:48:25 -07:00 |
|
rocm
|
[ROCm][Kernel] Add gfx950 support for skinny gemms (#18010)
|
2025-05-31 07:40:05 -07:00 |
|
sparse/cutlass
|
fix CUDA_check redefinition in #17918 (#18287)
|
2025-05-19 13:42:35 -07:00 |
|
activation_kernels.cu
|
Modularize fused experts and integrate PPLX kernels (#15956)
|
2025-05-14 13:11:54 -07:00 |
|
cache.h
|
[Attention] MLA with chunked prefill (#12639)
|
2025-02-21 15:30:12 -08:00 |
|
cache_kernels.cu
|
Allocate kv_cache with stride order (#16605)
|
2025-04-25 22:03:31 -07:00 |
|
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.h
|
[Attention] MLA with chunked prefill (#12639)
|
2025-02-21 15:30:12 -08:00 |
|
cuda_utils_kernels.cu
|
[NVIDIA] Support nvfp4 quantization (#12784)
|
2025-02-12 19:51:51 -08:00 |
|
cuda_view.cu
|
[V1] Fully Transparent Implementation of CPU Offloading (#15354)
|
2025-03-31 20:22:34 +08:00 |
|
cumem_allocator.cpp
|
[core] improve error handling when wake up from sleep mode (#12981)
|
2025-02-10 09:38:57 +08:00 |
|
custom_all_reduce.cu
|
[Distributed] Add custom allreduce support for ROCM (#14125)
|
2025-03-31 22:49:12 -07:00 |
|
custom_all_reduce.cuh
|
fix: spelling (#16466)
|
2025-04-11 23:24:22 -07:00 |
|
custom_all_reduce_test.cu
|
[Distributed] Add custom allreduce support for ROCM (#14125)
|
2025-03-31 22:49:12 -07:00 |
|
dispatch_utils.h
|
Modularize fused experts and integrate PPLX kernels (#15956)
|
2025-05-14 13:11:54 -07:00 |
|
layernorm_kernels.cu
|
[Kernel] Use fused rmsnorm for some models like qwen3 series (#17735)
|
2025-05-06 23:10:02 -07:00 |
|
layernorm_quant_kernels.cu
|
dynamic distpatch of fp8 kernels (#14245)
|
2025-03-11 10:54:56 -04:00 |
|
ops.h
|
[KERNEL] Sampler. CUDA kernel for applying repetition penalty (#18437)
|
2025-06-03 21:13:01 -07:00 |
|
permute_cols.cu
|
[Kernel] (2/N) Machete - Integrate into CompressedTensorsWNA16 and GPTQMarlin (#7701)
|
2024-09-23 13:46:26 -04:00 |
|
pos_encoding_kernels.cu
|
[Kernel] Have rotary embeddings support tensors (#18046)
|
2025-05-14 15:43:55 -07:00 |
|
sampler.cu
|
[KERNEL] Sampler. CUDA kernel for applying repetition penalty (#18437)
|
2025-06-03 21:13:01 -07:00 |
|
torch_bindings.cpp
|
[KERNEL] Sampler. CUDA kernel for applying repetition penalty (#18437)
|
2025-06-03 21:13:01 -07:00 |
|
type_convert.cuh
|
[torch.compile] Fuse RMSNorm with quant (#9138)
|
2024-11-08 21:20:08 +00:00 |
41aa578428