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

{
    "call": {
        "memory": {
            "unit": "MB",
            "max_ram": 1914.986496,
            "max_global_vram": 9371.643904,
            "max_process_vram": 1846670.266368,
            "max_reserved": 8923.38176,
            "max_allocated": 7787.456512
        },
        "latency": {
            "unit": "s",
            "count": 3,
            "total": 1.4216172485351564,
            "mean": 0.47387241617838544,
            "stdev": 0.005970492935751425,
            "p50": 0.47095220947265626,
            "p90": 0.4799453247070313,
            "p95": 0.4810694641113281,
            "p99": 0.48196877563476564,
            "values": [
                0.482193603515625,
                0.47095220947265626,
                0.468471435546875
            ]
        },
        "throughput": {
            "unit": "images/s",
            "value": 2.110272651159248
        },
        "energy": null,
        "efficiency": null
    }
}