xref: /aosp_15_r20/external/pytorch/benchmarks/inference/results/output_64_true.md (revision da0073e96a02ea20f0ac840b70461e3646d07c45) 
1## Batch Size 64 Compile true
2
3| Experiment | Warmup_latency (s) | Average_latency (s) | Throughput (samples/sec) | GPU Utilization (%) |
4| ---------- | ------------------ | ------------------- | ------------------------ | ------------------- |
5| original | 14.872 +/- 1.142 | 8.001 +/- 0.607 | 461.791 +/- 13.902 | 46.577 +/- 3.598 |
6| h2d_d2h_threads | 12.900 +/- 0.448 | 9.421 +/- 0.845 | 418.964 +/- 25.199 | 47.681 +/- 5.883 |
7