Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Matrix Multiplication on Synthetic Transformer Shapes Attention-Value Att ⊗ V
Loading...
6.38
Latency (µs)
BWTA_Att
-106.6848
656.5026
1,419.69
2,182.8774
Apr 5, 2026
Latency (µs)
Updated 11d ago
Evaluation Results
Method
Method
Links
Latency (µs)
BWTA_Att
Shape=Att: [128, 128],...
2026.04
6.38
BWTA_Att
Shape=Att: [512, 512],...
2026.04
9.77
cuBLAS
Shape=Att: [128, 128],...
2026.04
38.68
FP16
Shape=Att: [128, 128],...
2026.04
51.12
cuBLAS
Shape=Att: [512, 512],...
2026.04
57.42
FP16
Shape=Att: [512, 512],...
2026.04
145.9
BWTA_Att
Shape=Att: [2048, 2048...
2026.04
201.6
cuBLAS
Shape=Att: [2048, 2048...
2026.04
611.9
FP16
Shape=Att: [2048, 2048...
2026.04
2,833
Feedback
Search any
task
Search any
task
