Skip to content
KernelIndex
Search
Models
Records
GPUs
Docs
More
Records
GPUs
Docs
Sign in
Kernels
Serving
⏎
Kernels
Serving
⏎
20 operations match
Pick one to compare its implementations
Operation
Family
Published runs
Last run
Attention output projection with residual
other
110
2026-08-20
Basic transformer block
other
25
2026-08-22
Dual stream attention with conditional cross attention
other
24
2026-08-22
Joint attention context projection
other
27
2026-08-21
ConvNeXtV2 layer with NHWC persistence backward
other
32
2026-08-15
Residual coupling flow block
other
39
2026-08-23
Attention output reshape and projection
other
31
2026-08-18
Flash attention GQA ultralong
other
35
2026-08-20
Tanh gated residual add backward
other
23
2026-08-13
Attention value projection with transpose
other
32
2026-08-24
Cross attention QKV projection with GQA repeat backward
other
26
2026-08-22
ConvNeXtV2 block with GRN
other
22
2026-08-19
Fused statistics projection and split
other
31
2026-08-24
Altup predict correction cycle backward
other
29
2026-08-24
Conv2d residual block
other
30
2026-08-19
Hybrid attention mask preparation
other
31
2026-08-03
Strided conv2d downsampling
other
55
2026-08-24
Attention output projection with reshape backward
other
53
2026-08-16
Attention QKV projection with GQA reshape
other
40
2026-08-24
Attention weights matmul with value projection
other
26
2026-08-22
Ranked only against runs that measured the same thing.