| { | |
| "implementation": "gptoss_results", | |
| "config": { | |
| "warmup": 10, | |
| "iters": 50, | |
| "device": "cuda", | |
| "dtype": "torch.float32", | |
| "tokens": 100, | |
| "vary_inputs": true | |
| }, | |
| "stats": { | |
| "avg_ms": 45.23668440000279, | |
| "min_ms": 39.77574300000697, | |
| "max_ms": 49.24737599998252, | |
| "std_ms": 2.8509890347901194, | |
| "p50_ms": 46.46113100000093, | |
| "p95_ms": 48.43337355002859, | |
| "p99_ms": 49.07736706999401, | |
| "num_iters": 50, | |
| "tokens_per_s": 2210.595257507286, | |
| "throughput_variance": 143.00506086156602 | |
| }, | |
| "output_sum": 11.53223705291748 | |
| } |