Benchmark

AMSbench

What is AMSbench?

AMSbench is a benchmark of ~8000 questions to evaluate multi-modal LLMs on analog/mixed-signal circuit tasks like schematic recognition, analysis, and design.

Released: 2025-06-21
Modality: multimodal
Evaluates: 多模态, Multimodal, 理解, Understanding, 创作, Creation, AMS, Circuit, EDA, 科学智能, AI for Science, 图像理解, Image Understanding, 语言生成, Generation, 科学推理, Scientific Reasoning, 不支持, Unsupported
Openness: unknown
Importer: OpenCompass Hub
Reported scores: 0

Source provenance

Original evidence https://hub.opencompass.org.cn/dataset-detail/AMSbench

AMSbench paper, code and dataset

Paper https://arxiv.org/abs/2505.24138.pdf
Code https://github.com/Why0912/AMSBench
Dataset https://huggingface.co/datasets/wwhhyy/AMSBench

No reported scores are on record for this benchmark yet.

Which sources cite AMSbench?

AMSbench

Related 多模态, Multimodal, 理解, Understanding, 创作, Creation, AMS, Circuit, EDA, 科学智能, AI for Science, 图像理解, Image Understanding, 语言生成, Generation, 科学推理, Scientific Reasoning, 不支持, Unsupported benchmarks