cuda_inference_diffusers_text-to-image_CompVis/stable-diffusion-v1-4/benchmark_report.json

{
    "call": {
        "memory": {
            "unit": "MB",
            "max_ram": 1964.118016,
            "max_global_vram": 9371.783168,
            "max_process_vram": 2631543.537664,
            "max_reserved": 8923.38176,
            "max_allocated": 7787.456512
        },
        "latency": {
            "unit": "s",
            "count": 3,
            "total": 1.4602183532714843,
            "mean": 0.48673945109049477,
            "stdev": 0.00478242962605405,
            "p50": 0.48501498413085936,
            "p90": 0.49161527709960934,
            "p95": 0.4924403137207031,
            "p99": 0.4931003430175781,
            "values": [
                0.49326535034179686,
                0.48501498413085936,
                0.48193801879882814
            ]
        },
        "throughput": {
            "unit": "images/s",
            "value": 2.0544872575247237
        },
        "energy": null,
        "efficiency": null
    }
}