Mateusz Ozga fa3c6811d8 Disable conv for Filter1x1Stride1Pad0 when K or C is even (#2186)

2025-05-16 10:18:47 +02:00

..

block

Improve the general performance of the Preshuffled GEMM V3 & delete the unnecessary instances (#2166)

2025-05-12 09:52:58 -07:00

device

Disable conv for Filter1x1Stride1Pad0 when K or C is even (#2186)

2025-05-16 10:18:47 +02:00

element

Add grouped conv fwd bias relu instances (#2179)

2025-05-09 22:52:34 +02:00

grid

Switch to v2 pipeline for grouped conv bwd data (#2181)

2025-05-13 10:14:30 +02:00

thread

Moe gemm activation (#2026)

2025-04-23 10:35:34 +08:00

warp

MX GEMM - Add FP8 GEMM Tests for Different Layouts (#2152)

2025-05-01 11:55:48 -06:00