A model is still running

A model is still running or downloading. Leaving will stop the current operation. Stay on this page to let it finish.

Skip to content
LocalAI.toolsBeta

Join the waitlist

Get an email when the new apps and model runs launch.

No account needed.

MobileNet 3

Classify an image into 1,000 ImageNet categories locally using official TorchVision MobileNetV3 Small weights.

Desktop inferencePrivate · Runs on your computer
About this model

Classify an image into 1,000 ImageNet categories locally using official TorchVision MobileNetV3 Small weights.

Model released
Unknown
Source repository created
Nov 9, 2016
Model weights · FP32
10.2 MB

Best for

  • Sorting ordinary object photos
  • Learning and testing local image classification

Keep in mind

  • Classifies the whole image; no object boxes, identity recognition or image captions.
  • Only ImageNet categories; unfamiliar objects, small subjects, clutter and cropped images can be misclassified.
  • Confidence scores are relative softmax scores, not calibrated certainty; a blank or unfamiliar image still receives labels.
  • Do not use for medical, identity, safety or high-impact decisions.
  • Input is EXIF-oriented RGB then resized to shortest edge256 and center-cropped224; this can omit subjects near image edges. Alpha is composited over white.
  • The app downloads its own pinned CPython and isolated packages on request; installed Python, user packages and Python/pip configuration are not used. Automatic model-package provisioning is verified on Apple Silicon Mac; other platform stacks remain unverified.

Processed on your deviceYour input is processed locally in the desktop app. Model and runtime files download from external hosts.

Target platforms · verification required

  • macos · aarch64, x86_64
  • windows · x86_64

Recommended configuration

FP32 · cpu (verified platform only)

Sources & technical details
Original model
https://github.com/pytorch/vision
Release verification
Unknown
Release evidence
Exact original checkpoint public release date has not been verified. Repository creation, training, paper, artifact conversion and update dates are not release evidence.
Source updated
Jan 29, 2025
Native model files
https://download.pytorch.org/models/mobilenet_v3_small-047dcff4.pth
Files updated
Feb 8, 2021
Artifact revision
sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f

Desktop app

Sample result

MobileNetV3 Small — Image Classification

Run MobileNet 3 on your computer

Download the desktop app to load this model, provide input and view results locally.

Public desktop downloads are coming soon.

Device requirements

macOS Apple SiliconCPU / FP32Tested8.0 GiBRAM estimate
macOS Apple SiliconMPS / FP32Tested8.0 GiBRAM estimate
Windows Intel / AMDCPU / FP32Unverified8.0 GiBRAM estimate
macOS Intel / AMDCPU / FP32Unverified8.0 GiBRAM estimate
If your device falls short

MemoryAdvisory only: close other apps, reduce image/tile/token/batch settings or choose a smaller verified model/precision. CPU can need more RAM than GPU.

GPUIf the selected GPU runtime fails, existing CPU fallback may recover when enabled, but CPU has its own memory/runtime needs.

PlatformUse a compatible OS/architecture/runtime or another model. These estimates do not add hardware detection or block runs.

Benchmark evidence

Saved desktop measurements apply to the recorded hardware and runtime. Fallback cases report the device that actually ran; they do not prove GPU support on the requested device.

MobileNetV3 Small — Image Classification

63 configurations tested · 2 Windows backend summaries
Passed63
Failed0
TestedNov 1, 2026
Runtimepytorch 2.6.0

Quality criteria

{ "version": 1, "scope": "One independently identified Labrador photograph with mirror/downsize derivatives plus blank-image structural case; not an ImageNet validation set", "fixtures": [ { "id": "dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "identity", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "mirrored-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "mirror", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "small-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "resize112", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "blank", "transform": "white", "note": "Closed-set model still returns labels. Test finite normalized probabilities and dimensions, not a fabricated unknown class." } ], "cpuGpuMaxScoreError": 0.0001, "minimumRankedLabels": 5, "requirements": [ "finite scores in [0,1]", "sorted descending", "unique valid ImageNet class indices", "expected Labrador among top5 with score>=0.05 for dog derivatives", "full 1000-class scores sum to 1 within 1e-5 for blank" ] }

Windows rows show median warm inference and median load from fresh processes with cached files. Packaged Windows app verification is pending.

PrecisionDeviceResultPlatformInference / load
FP32CUDA · RTX 4070 TiAcceleratedPassedVery fastWindows 11 Pro · x86_644 ms / 2934 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch648 ms / 1118 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch648 ms / 1118 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch649 ms / 1102 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch649 ms / 1041 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch649 ms / 1102 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch649 ms / 1041 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch649 ms / 1118 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6410 ms / 1041 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6410 ms / 1041 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6410 ms / 1102 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6410 ms / 1118 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6410 ms / 1041 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6410 ms / 1102 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6410 ms / 1041 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6411 ms / 1102 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6411 ms / 1118 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6411 ms / 1102 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6411 ms / 1041 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6411 ms / 1041 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6411 ms / 1118 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6412 ms / 1118 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6412 ms / 1102 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6412 ms / 1102 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6413 ms / 1118 ms
FP32mpsAcceleratedPassedSlowmacos · aarch64209 ms / 1041 ms
FP32mpsAcceleratedPassedSlowmacos · aarch64246 ms / 1118 ms
FP32mpsAcceleratedPassedSlowmacos · aarch64248 ms / 1102 ms
FP32CPU · Ryzen 7 7700 (1 thread)PassedVery fastWindows 11 Pro · x86_645 ms / 2896 ms
FP32cpuPassedVery fastmacos · aarch649 ms / 978 ms
FP32cpuPassedVery fastmacos · aarch6410 ms / 978 ms
FP32cpuPassedVery fastmacos · aarch6410 ms / 978 ms
FP32cpuPassedVery fastmacos · aarch6410 ms / 978 ms
FP32cpuPassedVery fastmacos · aarch6410 ms / 978 ms
FP32cpuPassedVery fastmacos · aarch6410 ms / 978 ms
FP32cpuPassedVery fastmacos · aarch6410 ms / 978 ms
FP32cpuPassedVery fastmacos · aarch6411 ms / 978 ms
FP32cpuPassedVery fastmacos · aarch6411 ms / 978 ms
FP32cpuPassedVery fastmacos · aarch6411 ms / 1013 ms
FP32cpuPassedVery fastmacos · aarch6411 ms / 1013 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1013 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1015 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1013 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1013 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1015 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1015 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1013 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1015 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 986 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1015 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 986 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1015 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 986 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 986 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 986 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1015 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 986 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1013 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1013 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 986 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1015 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 986 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 986 ms
FP32cpuPassedVery fastmacos · aarch6412 ms / 1015 ms
FP32cpuPassedVery fastmacos · aarch6413 ms / 1013 ms
Test setup & sample details
Quality gate
{ "version": 1, "scope": "One independently identified Labrador photograph with mirror/downsize derivatives plus blank-image structural case; not an ImageNet validation set", "fixtures": [ { "id": "dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "identity", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "mirrored-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "mirror", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "small-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "resize112", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "blank", "transform": "white", "note": "Closed-set model still returns labels. Test finite normalized probabilities and dimensions, not a fabricated unknown class." } ], "cpuGpuMaxScoreError": 0.0001, "minimumRankedLabels": 5, "requirements": [ "finite scores in [0,1]", "sorted descending", "unique valid ImageNet class indices", "expected Labrador among top5 with score>=0.05 for dog derivatives", "full 1000-class scores sum to 1 within 1e-5 for blank" ] }
Tested
2026-11-01T11:15:26Z
Environment
{ "os": "macOS 27.2", "arch": "aarch64", "hardware": "Apple M3 Max", "scope": "One independently identified Labrador photo, mirror/downsize derivatives and blank structural case; not ImageNet accuracy certification" }
Runtime
pytorch 2.6.0
Artifact revision
sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f
Fixture
{
  "file": "model-samples/desktop/mobilenet-dog.jpg",
  "url": "https://assets.localai.tools/samples/v1/files/d4cb12ea8a768582562e1f1c0c930ea6fc4757f5b66551b986b41498efd75ca1/mobilenet-dog.jpg",
  "size": 83281,
  "sha1": "f1d042827c2f2154a6d8043b8e8f07c1be612941",
  "sha256": "d4cb12ea8a768582562e1f1c0c930ea6fc4757f5b66551b986b41498efd75ca1"
}
Rerun when
Artifact/runtime/preprocessing/config/host changes; Fixture checksum or frozen quality gate changes
Per-case quality & failure details
[
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1012.5654170001326,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 14.309458000070663,
      "inferenceMs": 11.558666999917477,
      "imagesPerSecond": 86.51516649862302,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1026.8748750002032,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1012.5654170001326,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.576542000009795,
      "inferenceMs": 12.054125000076965,
      "imagesPerSecond": 82.9591529865183,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1026.1419590001424,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1012.5654170001326,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.214500000003682,
      "inferenceMs": 11.412749999863081,
      "imagesPerSecond": 87.62130073926065,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1025.7799170001363,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1012.5654170001326,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 14.917249999825799,
      "inferenceMs": 13.425458000028812,
      "imagesPerSecond": 74.48535461493037,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1027.4826669999584,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1012.5654170001326,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.213624999934837,
      "inferenceMs": 11.660707999908482,
      "imagesPerSecond": 85.75808604484808,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1025.7790420000674,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1012.5654170001326,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.272124999957668,
      "inferenceMs": 11.699665999913123,
      "imagesPerSecond": 85.47252545563485,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1025.8375420000903,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.5261450409889221,
      "scoreSum": 0.9412240404635668,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1012.5654170001326,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.049874999978783,
      "inferenceMs": 11.525750000146218,
      "imagesPerSecond": 86.76224974403522,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1025.6152920001114,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.24970804154872894,
      "scoreSum": 0.955406092107296,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1012.5654170001326,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.831540999968638,
      "inferenceMs": 12.031417000116562,
      "imagesPerSecond": 83.11572942657642,
      "inputWidth": 112,
      "inputHeight": 92,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1025.3969580001012,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": null,
      "expectedScore": null,
      "scoreSum": 0.9999999155643309,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1012.5654170001326,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.534291000065423,
      "inferenceMs": 11.3774999999805,
      "imagesPerSecond": 87.89277081975072,
      "inputWidth": 224,
      "inputHeight": 224,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1025.099708000198,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1014.7253340001043,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.951500000075612,
      "inferenceMs": 12.17141699999047,
      "imagesPerSecond": 82.15970252278622,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1028.67683400018,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1014.7253340001043,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.486709000062547,
      "inferenceMs": 11.826041999938752,
      "imagesPerSecond": 84.5591449789523,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1028.212043000167,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1014.7253340001043,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.018166999927416,
      "inferenceMs": 11.537791999899127,
      "imagesPerSecond": 86.67169593703395,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1027.7435010000318,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1014.7253340001043,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.136499999973239,
      "inferenceMs": 11.686750000080792,
      "imagesPerSecond": 85.56698825533933,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1027.8618340000776,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1014.7253340001043,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.174208999998882,
      "inferenceMs": 11.775958000043829,
      "imagesPerSecond": 84.91878112984762,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1027.8995430001032,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1014.7253340001043,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.268292000020665,
      "inferenceMs": 11.66908400000466,
      "imagesPerSecond": 85.69652939336119,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1027.993626000125,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.5261450409889221,
      "scoreSum": 0.9412240404635668,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1014.7253340001043,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.82079099994371,
      "inferenceMs": 12.142790999860154,
      "imagesPerSecond": 82.35338976117738,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1028.546125000048,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.24970804154872894,
      "scoreSum": 0.955406092107296,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1014.7253340001043,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.486332999969818,
      "inferenceMs": 11.733416999959445,
      "imagesPerSecond": 85.22666500333673,
      "inputWidth": 112,
      "inputHeight": 92,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1027.2116670000742,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": null,
      "expectedScore": null,
      "scoreSum": 0.9999999155643309,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1014.7253340001043,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.192458000048646,
      "inferenceMs": 11.978666999993948,
      "imagesPerSecond": 83.48174300199724,
      "inputWidth": 224,
      "inputHeight": 224,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1027.917792000153,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 985.9926250001081,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.802250000026106,
      "inferenceMs": 12.158291999867288,
      "imagesPerSecond": 82.24839475897728,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 999.7948750001342,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 985.9926250001081,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.346917000035319,
      "inferenceMs": 11.738417000060508,
      "imagesPerSecond": 85.1903625501501,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 999.3395420001434,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 985.9926250001081,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.804499999878317,
      "inferenceMs": 12.144708000050741,
      "imagesPerSecond": 82.34039056318373,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 999.7971249999864,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 985.9926250001081,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.504125000054046,
      "inferenceMs": 11.91083300000173,
      "imagesPerSecond": 83.95718418685365,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 999.4967500001621,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 985.9926250001081,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.57029199994031,
      "inferenceMs": 12.017792000051486,
      "imagesPerSecond": 83.20996069791488,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 999.5629170000484,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 985.9926250001081,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.44362499980889,
      "inferenceMs": 11.811625000063941,
      "imagesPerSecond": 84.66235594125166,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 999.436249999917,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.5261450409889221,
      "scoreSum": 0.9412240404635668,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 985.9926250001081,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.583292000021174,
      "inferenceMs": 11.96183399997608,
      "imagesPerSecond": 83.59922065479255,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 999.5759170001293,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.24970804154872894,
      "scoreSum": 0.955406092107296,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 985.9926250001081,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.66079200013337,
      "inferenceMs": 11.926249999987704,
      "imagesPerSecond": 83.84865318109473,
      "inputWidth": 112,
      "inputHeight": 92,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 998.6534170002415,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": null,
      "expectedScore": null,
      "scoreSum": 0.9999999155643309,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 985.9926250001081,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.33937499998683,
      "inferenceMs": 12.139250000018365,
      "imagesPerSecond": 82.37741211347382,
      "inputWidth": 224,
      "inputHeight": 224,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 999.3320000000949,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 977.8839160001098,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 11.820791999980429,
      "inferenceMs": 10.164457999962906,
      "imagesPerSecond": 98.38202883062229,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 989.7047080000902,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 977.8839160001098,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 11.664917000189234,
      "inferenceMs": 10.247624999919935,
      "imagesPerSecond": 97.58358644152308,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 989.548833000299,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 977.8839160001098,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 11.495667000190224,
      "inferenceMs": 10.116417000062938,
      "imagesPerSecond": 98.84922695394808,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 989.3795830003,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 977.8839160001098,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.186832999987018,
      "inferenceMs": 10.674957999981416,
      "imagesPerSecond": 93.67718355442156,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 990.0707490000968,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 977.8839160001098,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.171916999932364,
      "inferenceMs": 10.770124999908148,
      "imagesPerSecond": 92.84943303894137,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 990.0558330000422,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.38916099071502686,
      "scoreSum": 0.8802293390035629,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 977.8839160001098,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 10.800959000107468,
      "inferenceMs": 9.282833000042956,
      "imagesPerSecond": 107.72573415845923,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 988.6848750002173,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.5261450409889221,
      "scoreSum": 0.9412240404635668,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 977.8839160001098,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 11.822624999922482,
      "inferenceMs": 10.364458000140075,
      "imagesPerSecond": 96.48357878303767,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 989.7065410000323,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.24970804154872894,
      "scoreSum": 0.955406092107296,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 977.8839160001098,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 10.989375000008295,
      "inferenceMs": 10.449375000007421,
      "imagesPerSecond": 95.69950355875731,
      "inputWidth": 112,
      "inputHeight": 92,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 988.8732910001181,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": null,
      "expectedScore": null,
      "scoreSum": 0.9999999155643309,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 977.8839160001098,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 11.204750000160857,
      "inferenceMs": 10.217750000037995,
      "imagesPerSecond": 97.86890460192132,
      "inputWidth": 224,
      "inputHeight": 224,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 989.0886660002707,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1041.2507089999963,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 344.0581669999574,
      "inferenceMs": 209.0159169999879,
      "imagesPerSecond": 4.784324631123944,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1385.3088759999537,
      "grade": "slow",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1041.2507089999963,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 14.731708999988768,
      "inferenceMs": 11.28266699993219,
      "imagesPerSecond": 88.63152657133372,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1055.982417999985,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1041.2507089999963,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 15.550332999964667,
      "inferenceMs": 10.308666000128142,
      "imagesPerSecond": 97.00576194704237,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1056.801041999961,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1041.2507089999963,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.904040999956123,
      "inferenceMs": 8.697333000100116,
      "imagesPerSecond": 114.97777536958615,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1054.1547499999524,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1041.2507089999963,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 15.19466700005978,
      "inferenceMs": 11.24883299985413,
      "imagesPerSecond": 88.89811058737982,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1056.445376000056,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1041.2507089999963,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 14.127833999964423,
      "inferenceMs": 9.117167000113113,
      "imagesPerSecond": 109.68319435056891,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1055.3785429999607,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.5261441469192505,
      "scoreSum": 0.9412234355695546,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1041.2507089999963,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.970624999956271,
      "inferenceMs": 9.751167000104033,
      "imagesPerSecond": 102.55182789806915,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1055.2213339999525,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.2497064471244812,
      "scoreSum": 0.955406004562974,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1041.2507089999963,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 14.037292000011803,
      "inferenceMs": 9.957207999832463,
      "imagesPerSecond": 100.4297590265088,
      "inputWidth": 112,
      "inputHeight": 92,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1055.288001000008,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": null,
      "expectedScore": null,
      "scoreSum": 1.000000001880835,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1041.2507089999963,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 17.48566700007359,
      "inferenceMs": 9.612625000045227,
      "imagesPerSecond": 104.02985656834579,
      "inputWidth": 224,
      "inputHeight": 224,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62914560,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1058.7363760000699,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1117.9074169999694,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 267.1287499999835,
      "inferenceMs": 245.954250000068,
      "imagesPerSecond": 4.065796789442441,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1385.036166999953,
      "grade": "slow",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1117.9074169999694,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 16.36729100005141,
      "inferenceMs": 10.939624999991793,
      "imagesPerSecond": 91.4108116138122,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1134.2747080000208,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1117.9074169999694,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.890458000129001,
      "inferenceMs": 8.31441599984828,
      "imagesPerSecond": 120.27302940077186,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1130.7978750000984,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1117.9074169999694,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 14.953791999914756,
      "inferenceMs": 11.310165999930177,
      "imagesPerSecond": 88.41603209061418,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1132.8612089998842,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1117.9074169999694,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.438750000001164,
      "inferenceMs": 8.45866700001352,
      "imagesPerSecond": 118.22193733343583,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1130.3461669999706,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1117.9074169999694,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.702916000056575,
      "inferenceMs": 9.887917000014568,
      "imagesPerSecond": 101.13353500019537,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1131.610333000026,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.5261441469192505,
      "scoreSum": 0.9412234355695546,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1117.9074169999694,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 17.212166999797773,
      "inferenceMs": 13.073250000161352,
      "imagesPerSecond": 76.49207350793857,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1135.1195839997672,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.2497064471244812,
      "scoreSum": 0.955406004562974,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1117.9074169999694,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 12.68079100009345,
      "inferenceMs": 9.491042000036032,
      "imagesPerSecond": 105.36250919511299,
      "inputWidth": 112,
      "inputHeight": 92,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1130.5882080000629,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": null,
      "expectedScore": null,
      "scoreSum": 1.000000001880835,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1117.9074169999694,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 18.073415999879217,
      "inferenceMs": 11.57941699989351,
      "imagesPerSecond": 86.36013367591792,
      "inputWidth": 224,
      "inputHeight": 224,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62914560,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1135.9808329998486,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1101.5959169999405,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 271.0092499999064,
      "inferenceMs": 248.15516700004991,
      "imagesPerSecond": 4.029736765464162,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1372.605166999847,
      "grade": "slow",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1101.5959169999405,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 15.438291000009485,
      "inferenceMs": 12.096500000097876,
      "imagesPerSecond": 82.66854048624882,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1117.03420799995,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1101.5959169999405,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 14.972207999790044,
      "inferenceMs": 10.574540999868987,
      "imagesPerSecond": 94.56675235477262,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1116.5681249997306,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1101.5959169999405,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 14.21541700005946,
      "inferenceMs": 8.616500000016458,
      "imagesPerSecond": 116.05640341183658,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1115.811334,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1101.5959169999405,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.164374999860229,
      "inferenceMs": 9.877000000187763,
      "imagesPerSecond": 101.24531740214537,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1114.7602919998008,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.3891615569591522,
      "scoreSum": 0.8802296463400126,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1101.5959169999405,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 15.145874999916487,
      "inferenceMs": 10.998416999882465,
      "imagesPerSecond": 90.92217543767312,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1116.741791999857,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.5261441469192505,
      "scoreSum": 0.9412234355695546,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1101.5959169999405,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 15.142333000085273,
      "inferenceMs": 11.674042000095142,
      "imagesPerSecond": 85.66013382441575,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1116.7382500000258,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": "Labrador retriever",
      "expectedScore": 0.2497064471244812,
      "scoreSum": 0.955406004562974,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1101.5959169999405,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 13.333874999943873,
      "inferenceMs": 10.018833000003724,
      "imagesPerSecond": 99.81202401513512,
      "inputWidth": 112,
      "inputHeight": 92,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62898176,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1114.9297919998844,
      "grade": "excellent",
      "fallback": false
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "expectedLabel": null,
      "expectedScore": null,
      "scoreSum": 1.000000001880835,
      "method": "Frozen known-object top5/ranking and blank normalization gates"
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 1101.5959169999405,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 14.803500000198255,
      "inferenceMs": 8.736666999993759,
      "imagesPerSecond": 114.46012535452185,
      "inputWidth": 224,
      "inputHeight": 224,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 62914560,
      "artifactRevision": "sha256:047dcff4addef86ea5bc2eff13c9614dc11f47ab1160d0a71a25e7db994f4e1f",
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 1116.3994170001388,
      "grade": "excellent",
      "fallback": false
    }
  }
]

MobileNetV3 Large · ImageNet1K V1

36 configurations tested
Passed36
Failed0
TestedNov 13, 2026
Runtimepytorch 2.6.0

Quality criteria

{ "version": 1, "scope": "One independently identified Labrador photograph with mirror/downsize derivatives plus blank-image structural case; not an ImageNet validation set", "fixtures": [ { "id": "dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "identity", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "mirrored-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "mirror", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "small-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "resize112", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "blank", "transform": "white", "note": "Closed-set model still returns labels. Test finite normalized probabilities and dimensions, not a fabricated unknown class." } ], "cpuGpuMaxScoreError": 0.0001, "minimumRankedLabels": 5, "requirements": [ "finite scores in [0,1]", "sorted descending", "unique valid ImageNet class indices", "expected Labrador among top5 with score>=0.05 for dog derivatives", "full 1000-class scores sum to 1 within 1e-5 for blank" ] }

PrecisionDeviceResultPlatformInference / load
FP32mpsAcceleratedPassedVery fastmacos · aarch6411 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6412 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6412 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6412 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6412 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6413 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6414 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6414 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6415 ms / 0 ms
FP32mpsAcceleratedPassedFastmacos · aarch64312 ms / 160 ms
FP32mpsAcceleratedPassedFastmacos · aarch64318 ms / 147 ms
FP32mpsAcceleratedPassedFastmacos · aarch64343 ms / 158 ms
FP32cpuPassedVery fastmacos · aarch6415 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6415 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6415 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6416 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6416 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6416 ms / 84 ms
FP32cpuPassedVery fastmacos · aarch6417 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6417 ms / 77 ms
FP32cpuPassedVery fastmacos · aarch6417 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6418 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6418 ms / 71 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 76 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6422 ms / 77 ms
FP32cpuPassedVery fastmacos · aarch6431 ms / 78 ms
FP32cpuPassedVery fastmacos · aarch6443 ms / 0 ms
Test setup & sample details
Quality gate
{ "version": 1, "scope": "One independently identified Labrador photograph with mirror/downsize derivatives plus blank-image structural case; not an ImageNet validation set", "fixtures": [ { "id": "dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "identity", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "mirrored-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "mirror", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "small-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "resize112", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "blank", "transform": "white", "note": "Closed-set model still returns labels. Test finite normalized probabilities and dimensions, not a fabricated unknown class." } ], "cpuGpuMaxScoreError": 0.0001, "minimumRankedLabels": 5, "requirements": [ "finite scores in [0,1]", "sorted descending", "unique valid ImageNet class indices", "expected Labrador among top5 with score>=0.05 for dog derivatives", "full 1000-class scores sum to 1 within 1e-5 for blank" ] }
Tested
2026-11-13T18:23:59.417563+00:00
Environment
{ "os": "macos", "arch": "aarch64", "hardware": "Apple M3 Max", "osVersion": "27.2", "python": "App-owned CPython3.13.18 private model probe venv" }
Runtime
pytorch 2.6.0
Artifact revision
sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04
Fixture
{
  "file": "model-samples/desktop/variants/mobilenet-large-v2/mobilenet-large-v2-dog.png",
  "size": 709899,
  "sha1": "409baef01589c0a55c12bba3b4f5330adc74b1c7",
  "sha256": "cc986ccf120f7892c9ee0202001c30a91a3039cc5859d851b981476507a25fe3",
  "url": "https://assets.localai.tools/samples/v1/files/cc986ccf120f7892c9ee0202001c30a91a3039cc5859d851b981476507a25fe3/mobilenet-large-v2-dog.png"
}
Rerun when
Artifact/runtime/preprocessing/config/host changes; Fixture checksum or frozen quality gate changes
Per-case quality & failure details
[
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 77.53229199443012,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 37.06875001080334,
      "inferenceMs": 30.87712498381734,
      "imagesPerSecond": 32.386434958698345,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 114.60104200523347,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 21.240334026515484,
      "inferenceMs": 19.643709005322307,
      "imagesPerSecond": 50.906883202609954,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.240334026515484,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 20.958083972800523,
      "inferenceMs": 20.172165997792035,
      "imagesPerSecond": 49.57325852411962,
      "inputWidth": 112,
      "inputHeight": 112,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 20.958083972800523,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": null,
      "fullProbabilitySum": 1.0000002196175046
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 21.137917006853968,
      "inferenceMs": 19.902166968677193,
      "imagesPerSecond": 50.245784872262355,
      "inputWidth": 229,
      "inputHeight": 317,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.137917006853968,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 21.148041996639222,
      "inferenceMs": 19.560499989893287,
      "imagesPerSecond": 51.12343756635515,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.148041996639222,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 21.20245900005102,
      "inferenceMs": 19.61499999742955,
      "imagesPerSecond": 50.98139179867678,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.20245900005102,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 76.40091597568244,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 22.139999957289547,
      "inferenceMs": 20.435833022929728,
      "imagesPerSecond": 48.93365486388368,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 98.54091593297198,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 22.110500023700297,
      "inferenceMs": 20.497541001532227,
      "imagesPerSecond": 48.78633978218403,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 22.110500023700297,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 21.787292032968253,
      "inferenceMs": 20.164207962807268,
      "imagesPerSecond": 49.59282317681372,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.787292032968253,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 76.88825001241639,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 23.517124995123595,
      "inferenceMs": 21.68916695518419,
      "imagesPerSecond": 46.10596626722807,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 100.40537500753999,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 21.890458010602742,
      "inferenceMs": 20.288917003199458,
      "imagesPerSecond": 49.28799303788887,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.890458010602742,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 21.593459008727223,
      "inferenceMs": 19.902166968677193,
      "imagesPerSecond": 50.245784872262355,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.593459008727223,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 76.6507500084117,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 18.65154196275398,
      "inferenceMs": 17.00779201928526,
      "imagesPerSecond": 58.796579759800245,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 95.30229197116569,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 16.204125015065074,
      "inferenceMs": 14.756542048417032,
      "imagesPerSecond": 67.76655375757713,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 16.204125015065074,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 17.390332999639213,
      "inferenceMs": 16.81591698434204,
      "imagesPerSecond": 59.467467693325275,
      "inputWidth": 112,
      "inputHeight": 112,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.390332999639213,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": null,
      "fullProbabilitySum": 1.0000002196175046
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 17.223625036422163,
      "inferenceMs": 16.142374952323735,
      "imagesPerSecond": 61.94875307713302,
      "inputWidth": 229,
      "inputHeight": 317,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.223625036422163,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 16.71395800076425,
      "inferenceMs": 15.245166956447065,
      "imagesPerSecond": 65.5945587776661,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 16.71395800076425,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 16.838125011418015,
      "inferenceMs": 15.424875018652529,
      "imagesPerSecond": 64.83034700707462,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 16.838125011418015,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 70.51037502242252,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 20.15741600189358,
      "inferenceMs": 18.488500034436584,
      "imagesPerSecond": 54.087676022251955,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 90.6677910243161,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 18.745750014204532,
      "inferenceMs": 17.170124978292733,
      "imagesPerSecond": 58.24069430270579,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 18.745750014204532,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 19.74058395717293,
      "inferenceMs": 18.21941597154364,
      "imagesPerSecond": 54.886501387413844,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 19.74058395717293,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 83.77854200080037,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 18.09208298800513,
      "inferenceMs": 16.433667042292655,
      "imagesPerSecond": 60.850691292847955,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 101.8706249888055,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 17.461624986026436,
      "inferenceMs": 15.94254100928083,
      "imagesPerSecond": 62.72525812653438,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.461624986026436,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 45.366167032625526,
      "inferenceMs": 43.34970802301541,
      "imagesPerSecond": 23.068206121920724,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 45.366167032625526,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 158.12737500527874,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 379.4165829895064,
      "inferenceMs": 343.2782089803368,
      "imagesPerSecond": 2.9130890742245765,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 537.5439579947852,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 16.667083953507245,
      "inferenceMs": 12.133790994994342,
      "imagesPerSecond": 82.41447379574436,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 16.667083953507245,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 14.243541983887553,
      "inferenceMs": 11.885292013175786,
      "imagesPerSecond": 84.13760460335521,
      "inputWidth": 112,
      "inputHeight": 112,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 14.243541983887553,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": null,
      "fullProbabilitySum": 1.0000000428335625
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 18.83341697975993,
      "inferenceMs": 11.493083031382412,
      "imagesPerSecond": 87.00885543673985,
      "inputWidth": 229,
      "inputHeight": 317,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84099072,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 18.83341697975993,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 16.8537920108065,
      "inferenceMs": 12.819207971915603,
      "imagesPerSecond": 78.00793950693412,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84099072,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 16.8537920108065,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 18.4294999926351,
      "inferenceMs": 14.0281249769032,
      "imagesPerSecond": 71.28536434102661,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84099072,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 18.4294999926351,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 146.97895903373137,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 346.722292015329,
      "inferenceMs": 317.7347920136526,
      "imagesPerSecond": 3.147278878911792,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 493.7012510490604,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 19.039458013139665,
      "inferenceMs": 14.38200002303347,
      "imagesPerSecond": 69.53135853139004,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 19.039458013139665,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 16.776874952483922,
      "inferenceMs": 12.161457969341427,
      "imagesPerSecond": 82.22698318910133,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 16.776874952483922,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 160.47250002156943,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 339.35104199917987,
      "inferenceMs": 311.61491700913757,
      "imagesPerSecond": 3.2090889922663,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 499.8235420207493,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 19.641083024907857,
      "inferenceMs": 14.572374988347292,
      "imagesPerSecond": 68.62299390453812,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 19.641083024907857,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "pythonVersion": "3.13.18",
      "pythonExecutable": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/bin/python",
      "pythonPrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv",
      "pythonBasePrefix": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/managed-runtime-models/.runtimes/cpython-3.13.18-20261026-aarch64-apple-darwin/python",
      "isolatedMode": true,
      "torchBuild": "2.6.0",
      "torchFile": "/Users/pengzhang/Desktop/hubters-webai/webai-js/localai-desktop/.cache/practical-variants/mobilenet-venv/lib/python3.13/site-packages/torch/__init__.py",
      "cudaRuntimeVersion": null,
      "cudaMatmulAllowTf32": null,
      "cudnnAllowTf32": null,
      "cudaDeviceName": null,
      "cudaComputeCapability": null,
      "durationMs": 16.380999993998557,
      "inferenceMs": 12.393624987453222,
      "imagesPerSecond": 80.68664341646269,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 16.380999993998557,
      "artifactRevision": "sha256:8738ca797c879b547d18bbd15da5736ff2557b2036a9af72225393ca61759a04",
      "fallback": false,
      "grade": "fast"
    }
  }
]

MobileNetV3 Large · ImageNet1K V2

36 configurations tested · 2 Windows backend summaries
Passed36
Failed0
TestedNov 1, 2026
Runtimepytorch 2.6.0

Quality criteria

{ "version": 1, "scope": "One independently identified Labrador photograph with mirror/downsize derivatives plus blank-image structural case; not an ImageNet validation set", "fixtures": [ { "id": "dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "identity", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "mirrored-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "mirror", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "small-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "resize112", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "blank", "transform": "white", "note": "Closed-set model still returns labels. Test finite normalized probabilities and dimensions, not a fabricated unknown class." } ], "cpuGpuMaxScoreError": 0.0001, "minimumRankedLabels": 5, "requirements": [ "finite scores in [0,1]", "sorted descending", "unique valid ImageNet class indices", "expected Labrador among top5 with score>=0.05 for dog derivatives", "full 1000-class scores sum to 1 within 1e-5 for blank" ] }

Windows rows show median warm inference and median load from fresh processes with cached files. Packaged Windows app verification is pending.

PrecisionDeviceResultPlatformInference / load
FP32CUDA · RTX 4070 TiAcceleratedPassedVery fastWindows 11 Pro · x86_645 ms / 3088 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6411 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6412 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6413 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6414 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6414 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6414 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6414 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6415 ms / 0 ms
FP32mpsAcceleratedPassedVery fastmacos · aarch6417 ms / 0 ms
FP32mpsAcceleratedPassedFastmacos · aarch64268 ms / 162 ms
FP32mpsAcceleratedPassedFastmacos · aarch64322 ms / 159 ms
FP32mpsAcceleratedPassedModeratemacos · aarch641824 ms / 161 ms
FP32CPU · Ryzen 7 7700 (1 thread)PassedVery fastWindows 11 Pro · x86_6410 ms / 3025 ms
FP32cpuPassedVery fastmacos · aarch6414 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6415 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6415 ms / 78 ms
FP32cpuPassedVery fastmacos · aarch6415 ms / 77 ms
FP32cpuPassedVery fastmacos · aarch6415 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6416 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6416 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6416 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6417 ms / 75 ms
FP32cpuPassedVery fastmacos · aarch6417 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6419 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6420 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6421 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6421 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6421 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6421 ms / 76 ms
FP32cpuPassedVery fastmacos · aarch6421 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6421 ms / 0 ms
FP32cpuPassedVery fastmacos · aarch6424 ms / 77 ms
FP32cpuPassedVery fastmacos · aarch6426 ms / 76 ms
Test setup & sample details
Quality gate
{ "version": 1, "scope": "One independently identified Labrador photograph with mirror/downsize derivatives plus blank-image structural case; not an ImageNet validation set", "fixtures": [ { "id": "dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "identity", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "mirrored-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "mirror", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "small-dog", "file": "model-samples/desktop/mobilenet-dog.jpg", "transform": "resize112", "expectedTop5": "Labrador retriever", "minimumExpectedScore": 0.05 }, { "id": "blank", "transform": "white", "note": "Closed-set model still returns labels. Test finite normalized probabilities and dimensions, not a fabricated unknown class." } ], "cpuGpuMaxScoreError": 0.0001, "minimumRankedLabels": 5, "requirements": [ "finite scores in [0,1]", "sorted descending", "unique valid ImageNet class indices", "expected Labrador among top5 with score>=0.05 for dog derivatives", "full 1000-class scores sum to 1 within 1e-5 for blank" ] }
Tested
2026-11-01T22:48:17.986373+00:00
Environment
{ "os": "macos", "arch": "aarch64", "hardware": "Apple M3 Max", "osVersion": "27.2", "python": "App-owned CPython3.13.18 private model probe venv" }
Runtime
pytorch 2.6.0
Artifact revision
sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997
Fixture
{
  "file": "model-samples/desktop/variants/mobilenet-large-v2/mobilenet-large-v2-dog.png",
  "size": 709899,
  "sha1": "409baef01589c0a55c12bba3b4f5330adc74b1c7",
  "sha256": "cc986ccf120f7892c9ee0202001c30a91a3039cc5859d851b981476507a25fe3",
  "url": "https://assets.localai.tools/samples/v1/files/cc986ccf120f7892c9ee0202001c30a91a3039cc5859d851b981476507a25fe3/mobilenet-large-v2-dog.png"
}
Rerun when
Artifact/runtime/preprocessing/config/host changes; Fixture checksum or frozen quality gate changes
Per-case quality & failure details
[
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 77.40466699760873,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 26.828500005649403,
      "inferenceMs": 24.267249995318707,
      "imagesPerSecond": 41.20780064460974,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 104.23316700325813,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 22.38020899676485,
      "inferenceMs": 20.6688749967725,
      "imagesPerSecond": 48.381926938749814,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 22.38020899676485,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 20.59591699799057,
      "inferenceMs": 19.885624999005813,
      "imagesPerSecond": 50.28758211270681,
      "inputWidth": 112,
      "inputHeight": 112,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 20.59591699799057,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": null,
      "fullProbabilitySum": 0.999999919149559
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 21.87916700495407,
      "inferenceMs": 20.639166999899317,
      "imagesPerSecond": 48.45156783725226,
      "inputWidth": 229,
      "inputHeight": 317,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.87916700495407,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 21.76958400377771,
      "inferenceMs": 20.023874996695668,
      "imagesPerSecond": 49.940383675238664,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.76958400377771,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 21.48624999972526,
      "inferenceMs": 19.824000002699904,
      "imagesPerSecond": 50.44390636923962,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.48624999972526,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 75.71558299969183,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 27.752709000196774,
      "inferenceMs": 25.838374996965285,
      "imagesPerSecond": 38.702124267391035,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 103.4682919998886,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 23.08712500234833,
      "inferenceMs": 21.23270800075261,
      "imagesPerSecond": 47.09714841670475,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 23.08712500234833,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 23.1799159955699,
      "inferenceMs": 21.454957997775637,
      "imagesPerSecond": 46.60927325533221,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 23.1799159955699,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 76.00541599822463,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 22.933500004000962,
      "inferenceMs": 21.133125002961606,
      "imagesPerSecond": 47.31907845431566,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 98.9389160022256,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 22.232916999200825,
      "inferenceMs": 20.6372079992434,
      "imagesPerSecond": 48.45616713446227,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 22.232916999200825,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 21.329582996258978,
      "inferenceMs": 19.77587500005029,
      "imagesPerSecond": 50.56666266334395,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 1,
      "totalMs": 21.329582996258978,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 74.84141700115288,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 18.190499999036547,
      "inferenceMs": 16.545042002690025,
      "imagesPerSecond": 60.44106747129515,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 93.03191700018942,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 20.239500001480337,
      "inferenceMs": 18.634000000020023,
      "imagesPerSecond": 53.6653429214836,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 20.239500001480337,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 20.556499999656808,
      "inferenceMs": 19.99062499817228,
      "imagesPerSecond": 50.023448496053966,
      "inputWidth": 112,
      "inputHeight": 112,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 20.556499999656808,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": null,
      "fullProbabilitySum": 0.999999919149559
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 17.85983300214866,
      "inferenceMs": 16.771375005191658,
      "imagesPerSecond": 59.625403384662576,
      "inputWidth": 229,
      "inputHeight": 317,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.85983300214866,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 17.608125002880115,
      "inferenceMs": 16.1080839970964,
      "imagesPerSecond": 62.08062983656264,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.608125002880115,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 16.776457996456884,
      "inferenceMs": 15.301833002013154,
      "imagesPerSecond": 65.35164773190486,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 16.776457996456884,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 77.70449999952689,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 16.769624999142252,
      "inferenceMs": 15.065666993905324,
      "imagesPerSecond": 66.37608546668001,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 94.47412499866914,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 17.48954100185074,
      "inferenceMs": 15.94370800012257,
      "imagesPerSecond": 62.72066698614352,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.48954100185074,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 17.31554200523533,
      "inferenceMs": 15.8755829979782,
      "imagesPerSecond": 62.98981272860045,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.31554200523533,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 76.5529999989667,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 16.757541998231318,
      "inferenceMs": 15.066916996147484,
      "imagesPerSecond": 66.37057868279847,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 93.31054199719802,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 16.321875002176967,
      "inferenceMs": 14.851291998638771,
      "imagesPerSecond": 67.33420904333826,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 16.321875002176967,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "cpu",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "placement": "cpu",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 15.175332999206148,
      "inferenceMs": 13.660791999427602,
      "imagesPerSecond": 73.2021979429817,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 15.175332999206148,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 161.3534999996773,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 1905.6532500035246,
      "inferenceMs": 1824.2081249991315,
      "imagesPerSecond": 0.5481830643641477,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84066304,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 2067.006750003202,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "moderate"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 18.438166996929795,
      "inferenceMs": 14.142333995550871,
      "imagesPerSecond": 70.70968627346777,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84066304,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 18.438166996929795,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 17.341833001410123,
      "inferenceMs": 13.425375000224449,
      "imagesPerSecond": 74.48581510634018,
      "inputWidth": 112,
      "inputHeight": 112,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84066304,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.341833001410123,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": null,
      "fullProbabilitySum": 1.0000000451691449
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 18.419583000650164,
      "inferenceMs": 11.934624999412335,
      "imagesPerSecond": 83.78981325757955,
      "inputWidth": 229,
      "inputHeight": 317,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 18.419583000650164,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 15.215749997878447,
      "inferenceMs": 11.289250003756024,
      "imagesPerSecond": 88.57984362710472,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 15.215749997878447,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 18.890499995904975,
      "inferenceMs": 13.97066699428251,
      "imagesPerSecond": 71.57854384541909,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 18.890499995904975,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 161.89254199707648,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 289.78179099794943,
      "inferenceMs": 268.19604099728167,
      "imagesPerSecond": 3.7286158150639355,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 451.6743329950259,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 20.843790996877942,
      "inferenceMs": 16.790000001492444,
      "imagesPerSecond": 59.55926145986368,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 20.843790996877942,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 17.307834001258016,
      "inferenceMs": 14.244375000998843,
      "imagesPerSecond": 70.20315036144991,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.307834001258016,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 158.9897910016589,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 347.030500000983,
      "inferenceMs": 321.86804200318875,
      "imagesPerSecond": 3.1068632778090253,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 506.0202910026419,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 19.01091599575011,
      "inferenceMs": 15.09829200222157,
      "imagesPerSecond": 66.2326573001012,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 19.01091599575011,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "passed": true,
      "validSortedUniquePredictions": true,
      "expectedLabradorTop5": true,
      "fullProbabilitySum": null
    },
    "runtime": {
      "engine": "pytorch",
      "version": "2.6.0",
      "torchvisionVersion": "0.21.0",
      "loadMs": 0,
      "device": "mps",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "placement": "mps:0",
      "parameterDtype": "torch.float32",
      "accelerationDetails": "Model and output on selected device; image decoding, EXIF, resize/crop/normalization and label assembly on CPU",
      "durationMs": 17.8714170033345,
      "inferenceMs": 13.920792000135407,
      "imagesPerSecond": 71.83499329565969,
      "inputWidth": 700,
      "inputHeight": 577,
      "modelInputShape": [
        1,
        3,
        224,
        224
      ],
      "driverAllocatedMemoryBytes": 84082688,
      "precision": "fp32",
      "cpuThreads": 4,
      "totalMs": 17.8714170033345,
      "artifactRevision": "sha256:5c1a416349c4cf298f2a6a5e2600ed0ee55e604713578f5e74e6bc8bcaef7997",
      "fallback": false,
      "grade": "fast"
    }
  }
]

Something not quite right?

Tell us what happened with MobileNet 3. Loading, results, or the page itself—we’re here for all of it.

Attachments Optional

Up to 3 files · 50 MB each

Browser, OS, CPU/GPU, memory, screen, storage, and network details are included automatically. Model inputs, results, and browsing history are not collected.