Richard Zou 76f44c2a8d Fix deepseek-v3 with torch.compile in PyTorch 2.6. (#5213)

2025-04-10 09:14:38 -07:00

..

allreduce

sgl-kernel transfer custom allreduce from trt kernel to vllm kernel (#5079)

2025-04-05 14:23:20 -07:00

attention

support cmake for sgl-kernel (#4706)

2025-03-27 01:42:28 -07:00

cpu

Add optimized native kernels in sgl-kernel (#5150)

2025-04-08 09:37:46 -07:00

cutlass_extensions

sgl-kernel use cutlass latest version for fp8 blockwise gemm (#5207)

2025-04-09 11:47:04 -07:00

elementwise

Optimize rope in sgl kernel (#4267)

2025-03-10 10:07:45 -07:00

gemm

sgl-kernel use cutlass latest version for fp8 blockwise gemm (#5207)

2025-04-09 11:47:04 -07:00

moe

reduce moe_align_block_size_kernel small batch mode overhead (#5086)

2025-04-09 17:59:35 -07:00

speculative

[ROCm] Enable MTP (NextN) on AMD GPU (#4631)

2025-03-23 22:58:05 -07:00

common_extension.cc

Fix deepseek-v3 with torch.compile in PyTorch 2.6. (#5213)

2025-04-10 09:14:38 -07:00

flash_extension.cc

[Fix] fix fa3 build at cu118 (#5036)

2025-04-03 11:52:35 -07:00

torch_extension_rocm.cc

[ROCm] Enable MTP (NextN) on AMD GPU (#4631)

2025-03-23 22:58:05 -07:00