Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multi-Head Attention Performance on NVIDIA L40S GPU (FP8, hd=128, Causal Mask)
Loading...
257.9
Performance (TFLOPS)
LLM-TL
223.476
232.413
241.35
250.287
Jun 14, 2025
Performance (TFLOPS)
Updated 4d ago
Evaluation Results
Method
Method
Links
Performance (TFLOPS)
LLM-TL
Sequence Length=16k, H...
2025.06
257.9
LLM-TL
Sequence Length=8k, He...
2025.06
255.1
LLM-TL
Sequence Length=4k, He...
2025.06
254.6
LLM-TL
Sequence Length=2k, He...
2025.06
248.3
LLM-TL
Sequence Length=1k, He...
2025.06
241.1
LLM-TL
Sequence Length=512, H...
2025.06
224.8
Feedback
Search any
task
Search any
task
