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

{
    "call": {
        "memory": {
            "unit": "MB",
            "max_ram": 1950.375936,
            "max_global_vram": 8310.4768,
            "max_process_vram": 2280036.974592,
            "max_reserved": 7862.222848,
            "max_allocated": 7785.1136
        },
        "latency": {
            "unit": "s",
            "count": 3,
            "total": 1.44374267578125,
            "mean": 0.48124755859374996,
            "stdev": 0.005815694337764941,
            "p50": 0.4799969482421875,
            "p90": 0.4871296142578125,
            "p95": 0.48802119750976564,
            "p99": 0.4887344641113281,
            "values": [
                0.48891278076171873,
                0.4748329467773437,
                0.4799969482421875
            ]
        },
        "throughput": {
            "unit": "images/s",
            "value": 2.0779326193822008
        },
        "energy": null,
        "efficiency": null
    }
}