# UnicomAI Benchmark

* [A-Eval](./A-Eval)：A benchmark designed to evaluate Chat LLMs of various scales from a practical application perspective.
* [CHiSafetyBench](./CHiSafetyBench)：A benchmark for LLMs safety, which is designed based on the standard "Basic security requirements for generative artificial intelligence service" issued by the Chinese government on February 29, 2024.
* [CDDMBench](./CDDMBench)：A multimodal benchmark dataset and model for crop disease diagnosis.
* [RAODBench](./RAODBench)：A Benchmark for Road Abandoned Object Detection from Video Surveillance.
* [TADBench](./TADBench)：A Large-scale Benchmark for Traffic Accidents Detection from Video Surveillance.
* [MITS](./Multimodal-Intelligent-Traffic-Surveillance)：A large-scale **multimodal dataset** for intelligent traffic surveillance.
## News
2024.6 We released an application-driven benchmark A-Eval.

2024.6 We released a Chinese safety benchmark CHiSafetyBench.

2024.7 We released A multimodal benchmark dataset for crop disease diagnosis, CDDMBench.

2024.12 We released a benchmark for road abandoned object detection from video surveillance, RAODBench.

2025.1 We released a large-scale benchmark for traffic accidents detection from video surveillance, TADBench.

2025.10 We released a large-scale **multimodal dataset** for intelligent traffic surveillance.
## Contact
China Unicom AI Innovation Center, China United Network Communication Group Co.,Ltd.