Codesota · Benchmark · MVTec ADHome/Leaderboards/MVTec AD
Unknown

MVTec AD.

5,354 high-resolution images across 15 object and texture categories. The gold standard for industrial anomaly detection with pixel-level annotations.

Paper ↗Leaderboard ↓
§ 01 · Leaderboard

Results by metric.

Found a wrong score or missing run?
Use row edits to send a sourced correction into moderation.
Add / edit result ↗Report issue ↗

Image AUROC

Image AUROC is the reported evaluation metric for MVTec AD. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Image AUROCverifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksEdit
01SimpleNet
CVPR 2023 SimpleNet, image-level AUROC across 15 categories
paper99.62026Source ↗Edit result
02PatchCore
Memory-bank based anomaly detection, widely-cited baseline
paper99.12026Source ↗Edit result
03EfficientAD
WACV 2024, real-time anomaly detection
paper99.12026Source ↗Edit result

auroc

Auroc is the reported evaluation metric for MVTec AD. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for aurocverifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksEdit
01simplenet
State-of-the-art on MVTec AD. Simple architecture.
paper99.62025Source ↗Edit result
02FastFlow
2D normalizing flow. Good speed-accuracy tradeoff.
paper99.42025Source ↗Edit result
03PatchCore
Image-level AUROC. State-of-the-art with WideResNet-50 backbone.
paper99.12025Source ↗Edit result
04efficientad
Matches PatchCore with 614 FPS inference.
paper99.12025Source ↗Edit result
05reverse-distillation
Knowledge distillation approach.
paper98.52025Source ↗Edit result
06Reverse Distillation
Knowledge distillation approach.
paper98.52025Source ↗Edit result
07CFLOW-AD
Conditional normalizing flows.
paper98.32025Source ↗Edit result
08DRAEM
Discriminatively trained reconstruction.
paper982025Source ↗Edit result
09padim
ResNet-18 backbone. Fast inference.
paper97.92025Source ↗Edit result
10AnomalyGPT
Image-level AUROC, unsupervised (zero-shot) setting. Table 1 in paper.
verified97.42023Paper ↗Edit result
§ 04 · Submit a result

Add to the leaderboard.

← Back to Leaderboards