AMSbench

该数据集是一个专为评估多模态大型语言模型（MLLM）在关键任务上的性能而设计的基准测试套件，这些任务包括电路原理图感知、电路分析和电路设计。该数据集包含大约8000个测试问题，覆盖了多个难度级别。此外，该基准测试评估了八种知名模型，包括开源和专有解决方案。规模上，数据集大约有8000个测试问题，其任务重点是评估MLLM在模拟电路任务上的性能。

This dataset is a benchmark suite specifically designed to evaluate the performance of Multimodal Large Language Models (MLLMs) on core tasks including circuit schematic perception, circuit analysis, and circuit design. It contains approximately 8,000 test questions spanning multiple difficulty levels. Additionally, this benchmark evaluates eight well-known models, including both open-source and proprietary solutions. In terms of scale, the dataset has approximately 8,000 test questions, and its tasks focus on assessing the performance of MLLMs on analog circuit tasks.