| { | |
| "implementation": "gptoss_results", | |
| "config": { | |
| "warmup": 10, | |
| "iters": 50, | |
| "device": "cuda", | |
| "dtype": "torch.float32", | |
| "tokens": 100, | |
| "vary_inputs": true | |
| }, | |
| "stats": { | |
| "avg_ms": 45.01105025997276, | |
| "min_ms": 39.02894699967874, | |
| "max_ms": 49.29527800004507, | |
| "std_ms": 2.979711623110132, | |
| "p50_ms": 45.6719464998514, | |
| "p95_ms": 48.48902935004844, | |
| "p99_ms": 49.0557057300839, | |
| "num_iters": 50, | |
| "tokens_per_s": 2221.6766643396363, | |
| "throughput_variance": 151.30753386326467 | |
| }, | |
| "output_sum": 11.53223705291748 | |
| } |