Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-Head Self-Attention on CUDA-LLM kernels task suite (test)
Loading...
-
Latency (s)
No plottable results for Latency (s) (TIME).
Metric
Latency (s) (TIME)
Speedup (SCALAR)
SM Compute Utilization (PERCENT)
DRAM Memory Utilization (PERCENT)
Texture Utilization (PERCENT)
DRAM Bytes Transferred (SCALAR)
L1 Sectors Accessed (SCALAR)
Instructions Executed (SCALAR)
Updated 5d ago
Evaluation Results
Method
Method
Links
Latency (s)
Speedup
SM Compute Utilization
DRAM Memory Utilization
Texture Utilization
DRAM Bytes Transferred
L1 Sectors Accessed
Instructions Executed
No evaluation results found.
Feedback
Search any
task
Search any
task
