Multi-sensor calibration precision is critical for the environmental perception and decision-making capabilities of robotic systems, particularly for the primary sensors such as LiDAR and cameras. However, current calibration efforts suffer from a lack of standardized benchmarks and typically focus on specific sensors, leading to varied experimental setups and a shortage of universally applicable evaluation metrics. Therefore, we propose a Benchmark for multi-LiDAR and multi-Camera Calibration (BLCC). Firstly, we design a set of evaluation metrics, using angle error and distance error, to assess the results of arbitrary calibration for cameras and LiDARs (multi-LiDAR, multi-camera, or LiDAR-camera). Specifically, multi-sensor calibration results consist of rotation and translation matrices. The angle error evaluates the precision of the rotation matrix, while the distance error evaluates the precision of the translation matrix. This dual-metric ensures comprehensive assessment of calibration quality. Additionally, we develop a multi-sensor calibration dataset to support the application of our proposed metrics. The dataset includes 4,848 images for multi-camera calibration, 352 images and 352 point cloud frames for camera-LiDAR calibration, and 1,296 point cloud frames for multi-LiDAR calibration. Finally, we conduct extensive experiments using BLCC to validate the performance of several widely recognized classical calibration methods. The code and dataset are available at: https://anonymous.4open.science/r/BLCC-2B4F .