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

{
    "call": {
        "memory": {
            "unit": "MB",
            "max_ram": 1957.82656,
            "max_global_vram": 8309.907456,
            "max_process_vram": 2304395.259904,
            "max_reserved": 7862.222848,
            "max_allocated": 7785.1136
        },
        "latency": {
            "unit": "s",
            "count": 3,
            "total": 1.4345322570800783,
            "mean": 0.47817741902669275,
            "stdev": 0.006413687538136851,
            "p50": 0.4820821228027344,
            "p90": 0.48306847534179687,
            "p95": 0.4831917694091797,
            "p99": 0.4832904046630859,
            "values": [
                0.4833150634765625,
                0.4820821228027344,
                0.46913507080078126
            ]
        },
        "throughput": {
            "unit": "images/s",
            "value": 2.09127399205812
        },
        "energy": null,
        "efficiency": null
    }
}