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

{
    "call": {
        "memory": {
            "unit": "MB",
            "max_ram": 1950.277632,
            "max_global_vram": 8309.972992,
            "max_process_vram": 2320754.774016,
            "max_reserved": 7862.222848,
            "max_allocated": 7785.1136
        },
        "latency": {
            "unit": "s",
            "count": 3,
            "total": 1.4413817749023439,
            "mean": 0.48046059163411464,
            "stdev": 0.007351992723843892,
            "p50": 0.4807897644042969,
            "p90": 0.4875945983886719,
            "p95": 0.48844520263671876,
            "p99": 0.48912568603515627,
            "values": [
                0.48929580688476565,
                0.4807897644042969,
                0.47129620361328123
            ]
        },
        "throughput": {
            "unit": "images/s",
            "value": 2.0813361541242292
        },
        "energy": null,
        "efficiency": null
    }
}