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

{
    "call": {
        "memory": {
            "unit": "MB",
            "max_ram": 1951.080448,
            "max_global_vram": 9400.91392,
            "max_process_vram": 2621321.728,
            "max_reserved": 8942.256128,
            "max_allocated": 7783.72096
        },
        "latency": {
            "unit": "s",
            "count": 3,
            "total": 1.4512677001953125,
            "mean": 0.48375590006510416,
            "stdev": 0.006969397217332115,
            "p50": 0.488598388671875,
            "p90": 0.48873498535156246,
            "p95": 0.48875205993652343,
            "p99": 0.48876571960449217,
            "values": [
                0.48876913452148435,
                0.488598388671875,
                0.47390017700195314
            ]
        },
        "throughput": {
            "unit": "images/s",
            "value": 2.067158250401534
        },
        "energy": null,
        "efficiency": null
    }
}