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

{
    "call": {
        "memory": {
            "unit": "MB",
            "max_ram": 1955.30752,
            "max_global_vram": 9423.314944,
            "max_process_vram": 2636426.887168,
            "max_reserved": 8952.741888,
            "max_allocated": 7781.492736
        },
        "latency": {
            "unit": "s",
            "count": 3,
            "total": 1.4632354431152343,
            "mean": 0.4877451477050781,
            "stdev": 0.0027497795953171787,
            "p50": 0.48894338989257813,
            "p90": 0.4900686340332031,
            "p95": 0.49020928955078125,
            "p99": 0.49032181396484376,
            "values": [
                0.48894338989257813,
                0.4903499450683594,
                0.4839421081542969
            ]
        },
        "throughput": {
            "unit": "images/s",
            "value": 2.050251047509475
        },
        "energy": null,
        "efficiency": null
    }
}