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

{
    "call": {
        "memory": {
            "unit": "MB",
            "max_ram": 1327.263744,
            "max_global_vram": 8143.765504,
            "max_process_vram": 0.0,
            "max_reserved": 7488.929792,
            "max_allocated": 6529.611776
        },
        "latency": {
            "unit": "s",
            "count": 2,
            "total": 1.3028583984375,
            "mean": 0.65142919921875,
            "stdev": 0.0043214111328124805,
            "p50": 0.65142919921875,
            "p90": 0.6548863281249999,
            "p95": 0.6553184692382812,
            "p99": 0.6556641821289062,
            "values": [
                0.6557506103515625,
                0.6471077880859375
            ]
        },
        "throughput": {
            "unit": "images/s",
            "value": 1.5350862399156902
        },
        "energy": {
            "unit": "kWh",
            "cpu": 7.679569472869238e-06,
            "ram": 4.191725337898096e-06,
            "gpu": 4.3518368148001e-05,
            "total": 5.5389662958768335e-05
        },
        "efficiency": {
            "unit": "images/kWh",
            "value": 18053.91018075688
        }
    }
}