The Codes and Data of The First-Ever Comprehensive Benchmark for Multimodal Large Language Models in Industrial Anomaly Detection
active 2024-09-23 → 2026-05-23 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 608 days from 2024-09-23 to 2026-05-23. Pushes: 21 total, peak 2 in a day. Pull requests: 2 total, peak 1 in a day. Issues: 31 total, peak 4 in a day. Comments: 18 total, peak 3 in a day. Stars: 173 total, peak 5 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| jam-cc | 45 | 21 | 1 | 10 |
| XiangnanHe | 3 | 0 | 1 | 2 |
| leejq666 | 2 | 0 | 0 | 0 |
| San-dow | 2 | 0 | 0 | 1 |
| jjhhyyg | 2 | 0 | 0 | 0 |
| pipixin321 | 2 | 0 | 0 | 0 |
| SeungdongYoa | 2 | 0 | 0 | 1 |
| wangerniuniu | 2 | 0 | 0 | 0 |
| LGRlgr6666 | 2 | 0 | 0 | 0 |
| zqhang | 1 | 0 | 0 | 0 |
| cvtower | 1 | 0 | 0 | 0 |
| KomorebiWerts | 1 | 0 | 0 | 0 |
| zhaoyuan1209 | 1 | 0 | 0 | 1 |
| Linda6521 | 1 | 0 | 0 | 0 |
| oysterhyd | 1 | 0 | 0 | 0 |
| caoyunkang | 1 | 0 | 0 | 1 |
| nicebro123 | 1 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#20jam-cc2026-01-14 10:58Title: [Bug/Data Quality] Systematic Logic Errors in "Defect Localization" GT for 'Misplaced' Category (Transistor)
- Issue#18jam-cc2026-01-14 10:56Defect Localization
- Issue#22oysterhyd2025-12-26 16:19Defect Localization found that a large number of answers to the questions are incorrect.
- Issue#21jjhhyyg2025-12-24 13:57DS-MVTec Dataset Photos Classification Problem
- Issue#18KomorebiWerts2025-12-02 15:39Defect Localization
- Issue#17jam-cc2025-11-17 09:40您好，请问完整代码什么时候开源？
- Issue#17leejq6662025-11-13 07:38您好，请问完整代码什么时候开源？
- Issue#16jam-cc2025-11-11 06:17how to use mmad to evaluate anomalyclip's performance in 7 subtasks?
- Issue comment#16jam-cc2025-11-11 06:17how to use mmad to evaluate anomalyclip's performance in 7 subtasks?
- Issue#16Linda65212025-11-09 11:22how to use mmad to evaluate anomalyclip's performance in 7 subtasks?
- Issue#15jam-cc2025-08-06 10:21Error for directory: 'ALL_DATA\DS-MVTec\pill\image\good'
- Issue comment#15jam-cc2025-07-21 07:52Error for directory: 'ALL_DATA\DS-MVTec\pill\image\good'
- Issue#15guanwei492025-07-21 05:33error for dir: 'ALL_DATA\DS-MVTec\pill\image\good'
- Issue#13jam-cc2025-07-14 13:44Code operation issue
- Issue#14jam-cc2025-07-08 10:00Qwen2.5-VL support of qwen_query.py
- Issue comment#14jam-cc2025-07-08 10:00Qwen2.5-VL support of qwen_query.py
- Issue#12jam-cc2025-07-08 03:29Is the dataset too large to download? I see that the dataset on huggingface is a form of description. I can't see where the four datasets you mentioned are, author
- Issue#14jjhhyyg2025-07-08 02:55Qwen2.5-VL support of qwen_query.py
- Issue#13LGRlgr66662025-07-04 07:25Code operation issue
- Issue#12LGRlgr66662025-07-03 08:21Is the dataset too large to download? I see that the dataset on huggingface is a form of description. I can't see where the four datasets you mentioned are, author
- Issue#8jam-cc2025-05-26 11:25Request more detailed test results
- Issue comment#11San-dow2025-05-26 09:58"cambrian_query.py" Check if there are any errors in the implementation code.
- Issue#11San-dow2025-05-26 07:45"cambrian_query.py" Check if there are any errors in the implementation code.
- Issue#10jam-cc2025-05-26 07:36您好！可以询问一下大模型所需的硬件条件吗？
- Issue comment#10jam-cc2025-05-26 07:36您好！可以询问一下大模型所需的硬件条件吗？
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 173 stars here means stars gained during the window, not the repo's star count.