# Industrial Anomaly Detection on MMAD one-shot 1.0 (test) Source: [Industrial Anomaly Detection on MMAD one-shot 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:34:56.055Z Evaluation of Industrial Anomaly Detection performance on the test split of the MMAD one-shot 1.0 dataset, reporting metrics across discrimination, classification, localization, and analysis. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Industrial Anomaly Detection | | Dataset (including any stated split) | MMAD one-shot 1.0 (test) | | Metric | Anomaly Discrimination Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 29 of 29 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Anomaly Discrimination Score](): Higher is better; unit: PERCENT. - [Anomaly Classification Score](): Higher is better; unit: PERCENT. - [Defect Localization Score](): Higher is better; unit: PERCENT. - [Defect Description Score](): Higher is better; unit: PERCENT. - [Defect Analysis Score](): Higher is better; unit: PERCENT. - [Object Classification Score](): Higher is better; unit: PERCENT. - [Object Analysis Score](): Higher is better; unit: PERCENT. - [Average Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Anomaly Discrimination Score | Result date | Report paper | | --- | --- | --- | --- | | Human (expert) | 95.24 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Human (ordinary) | 86.9 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Gemini-1.5-flash | 83.38 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Gemini-1.5-pro | 83.07 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Reason-IAD (Qwen3-VL) | 75.24 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Reason-IAD (Qwen3-VL) | 75.05 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Qwen3-VL | 74.02 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Qwen3-VL | 73.08 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | GLM-4.1V-Thinking | 72.31 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Qwen2.5-VL | 71.79 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Reason-IAD (Qwen2.5-VL) | 70.74 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Qwen3-VL | 70.29 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | GPT-4o | 68.63 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | InternVL2 | 68.25 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Kimi-VL | 66.57 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | AnomalyGPT | 65.57 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | GPT-4o-mini | 64.1 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Qwen2.5-VL | 63.03 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Reason-IAD (Qwen3-VL) | 61.9 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | AnomalyR1 | 60.62 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Claude-3.5-sonnet | 60.14 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | InternVL2 | 59.97 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | LLaVA-NeXT | 57.92 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | LLaVA-NeXT | 57.64 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Qwen3-VL-thinking | 52.31 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | LLaVA-OneVision | 51.77 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | LLaVA-1.5 | 51.33 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | Random Chance | 50 | Not recorded | 698c0d22c97a0b3e3b5e11cf | | LLaVA-1.5 | 49.96 | Not recorded | 698c0d22c97a0b3e3b5e11cf | ### Human (expert) Anomaly Discrimination Score: 95.24. Result ID: 69971bb55698665ec469b3bc. [Reporting paper]() Recorded evaluation configuration: {"Type":"Human"}. This record does not assert independent reproduction. ### Human (ordinary) Anomaly Discrimination Score: 86.9. Result ID: 69971bb55698665ec469b3bd. [Reporting paper]() Recorded evaluation configuration: {"Type":"Human"}. This record does not assert independent reproduction. ### Gemini-1.5-flash Anomaly Discrimination Score: 83.38. Result ID: 69971bb55698665ec469b3c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Commercial"}. This record does not assert independent reproduction. ### Gemini-1.5-pro Anomaly Discrimination Score: 83.07. Result ID: 69971bb55698665ec469b3bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Commercial"}. This record does not assert independent reproduction. ### Reason-IAD (Qwen3-VL) Anomaly Discrimination Score: 75.24. Result ID: 69971bb55698665ec469b3d7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Reason-IAD","Backbone":"Qwen3-VL","Scale":"8B"}. This record does not assert independent reproduction. ### Reason-IAD (Qwen3-VL) Anomaly Discrimination Score: 75.05. Result ID: 69971bb55698665ec469b3d5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Reason-IAD","Backbone":"Qwen3-VL","Scale":"4B"}. This record does not assert independent reproduction. ### Qwen3-VL Anomaly Discrimination Score: 74.02. Result ID: 69971bb55698665ec469b3c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"4B"}. This record does not assert independent reproduction. ### Qwen3-VL Anomaly Discrimination Score: 73.08. Result ID: 69971bb55698665ec469b3cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"8B"}. This record does not assert independent reproduction. ### GLM-4.1V-Thinking Anomaly Discrimination Score: 72.31. Result ID: 69971bb55698665ec469b3cf. [Reporting paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"9B"}. This record does not assert independent reproduction. ### Qwen2.5-VL Anomaly Discrimination Score: 71.79. Result ID: 69971bb55698665ec469b3c8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"7B"}. This record does not assert independent reproduction. ### Reason-IAD (Qwen2.5-VL) Anomaly Discrimination Score: 70.74. Result ID: 69971bb55698665ec469b3d6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Reason-IAD","Backbone":"Qwen2.5-VL","Scale":"7B"}. This record does not assert independent reproduction. ### Qwen3-VL Anomaly Discrimination Score: 70.29. Result ID: 69971bb55698665ec469b3c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"2B"}. This record does not assert independent reproduction. ### GPT-4o Anomaly Discrimination Score: 68.63. Result ID: 69971bb55698665ec469b3c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Commercial"}. This record does not assert independent reproduction. ### InternVL2 Anomaly Discrimination Score: 68.25. Result ID: 69971bb55698665ec469b3d3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"76B"}. This record does not assert independent reproduction. ### Kimi-VL Anomaly Discrimination Score: 66.57. Result ID: 69971bb55698665ec469b3d1. [Reporting paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"16B"}. This record does not assert independent reproduction. ### AnomalyGPT Anomaly Discrimination Score: 65.57. Result ID: 69971bb55698665ec469b3c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"7B"}. This record does not assert independent reproduction. ### GPT-4o-mini Anomaly Discrimination Score: 64.1. Result ID: 69971bb55698665ec469b3c2. [Reporting paper]() Recorded evaluation configuration: {"Type":"Commercial"}. This record does not assert independent reproduction. ### Qwen2.5-VL Anomaly Discrimination Score: 63.03. Result ID: 69971bb55698665ec469b3c4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"3B"}. This record does not assert independent reproduction. ### Reason-IAD (Qwen3-VL) Anomaly Discrimination Score: 61.9. Result ID: 69971bb55698665ec469b3d4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Reason-IAD","Backbone":"Qwen3-VL","Scale":"2B"}. This record does not assert independent reproduction. ### AnomalyR1 Anomaly Discrimination Score: 60.62. Result ID: 69971bb55698665ec469b3c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"3B"}. This record does not assert independent reproduction. ### Claude-3.5-sonnet Anomaly Discrimination Score: 60.14. Result ID: 69971bb55698665ec469b3be. [Reporting paper]() Recorded evaluation configuration: {"Type":"Commercial"}. This record does not assert independent reproduction. ### InternVL2 Anomaly Discrimination Score: 59.97. Result ID: 69971bb55698665ec469b3cc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"8B"}. This record does not assert independent reproduction. ### LLaVA-NeXT Anomaly Discrimination Score: 57.92. Result ID: 69971bb55698665ec469b3d2. [Reporting paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"34B"}. This record does not assert independent reproduction. ### LLaVA-NeXT Anomaly Discrimination Score: 57.64. Result ID: 69971bb55698665ec469b3ca. [Reporting paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"7B"}. This record does not assert independent reproduction. ### Qwen3-VL-thinking Anomaly Discrimination Score: 52.31. Result ID: 69971bb55698665ec469b3ce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"8B"}. This record does not assert independent reproduction. ### LLaVA-OneVision Anomaly Discrimination Score: 51.77. Result ID: 69971bb55698665ec469b3cb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"7B"}. This record does not assert independent reproduction. ### LLaVA-1.5 Anomaly Discrimination Score: 51.33. Result ID: 69971bb55698665ec469b3c9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"7B"}. This record does not assert independent reproduction. ### Random Chance Anomaly Discrimination Score: 50. Result ID: 69971bb55698665ec469b3bb. [Reporting paper]() Recorded evaluation configuration: {"Type":"Random"}. This record does not assert independent reproduction. ### LLaVA-1.5 Anomaly Discrimination Score: 49.96. Result ID: 69971bb55698665ec469b3d0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Type":"Open Source","Scale":"13B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.