Yashvardhan Agarwal 7f14772406 CK_TILE: Implement two-stage split-K GEMM with workspace reduction (LWPCK-2966) (#2632) ...

* CK_TILE: Implement two-stage split-K GEMM with reduction

- Added split-K GEMM with reduction example

* comment resolutions

2025-08-14 10:18:52 +02:00

..

01_fmha

fix for aiter consume (#2677)

2025-08-13 19:06:22 +08:00

02_layernorm2d

…

03_gemm

CK_TILE: Implement two-stage split-K GEMM with workspace reduction (LWPCK-2966) (#2632)

2025-08-14 10:18:52 +02:00

04_img2col

…

05_reduce

…

06_permute

…

09_topk_softmax

…

10_rmsnorm2d

…

11_add_rmsnorm2d_rdquant

…

12_smoothquant

…

13_moe_sorting

…

14_moe_smoothquant

…

15_fused_moe

…

16_batched_gemm

…

17_grouped_gemm

Finish the grouped gemm restructure with fp8 data type (#2655)

2025-08-12 18:23:34 -07:00

18_flatmm

Jimniu/tile_example_flatmm_basic fix (#2680)

2025-08-13 16:06:08 -07:00

19_gemm_multi_d

…

20_grouped_convolution

…

21_elementwise

[CK_TILE]fix elementwise example in gfx11/12 (#2676)

2025-08-13 15:21:46 -07:00

35_batched_transpose

…

38_block_scale_gemm

Preshuffle AQ matrix in block scale gemm (#2624)

2025-08-12 21:32:51 -07:00

39_copy

Minor Improvements in CK TILE memory copy EXAMPLE (#2678)

2025-08-13 15:24:16 -07:00

CMakeLists.txt

…

remod.py

…