Saved in:
Bibliographic Details
Main Authors: Lai, Yao, Poddar, Souradip, Lee, Sungyoung, Chen, Guojin, Hu, Mengkang, Yu, Bei, Luo, Ping, Pan, David Z.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.02518
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908511131140096
author Lai, Yao
Poddar, Souradip
Lee, Sungyoung
Chen, Guojin
Hu, Mengkang
Yu, Bei
Luo, Ping
Pan, David Z.
author_facet Lai, Yao
Poddar, Souradip
Lee, Sungyoung
Chen, Guojin
Hu, Mengkang
Yu, Bei
Luo, Ping
Pan, David Z.
contents Despite recent advances, analog front-end design still relies heavily on expert intuition and iterative simulations, which limits the potential for automation. We present AnalogCoder-Pro, a multimodal large language model (LLM) framework that integrates generative and optimization techniques. The framework features a multimodal diagnosis-and-repair feedback loop that uses simulation error messages and waveform images to autonomously correct design errors. It also builds a reusable circuit tool library by archiving successful designs as modular subcircuits, accelerating the development of complex systems. Furthermore, it enables end-to-end automation by generating circuit topologies from target specifications, extracting key parameters, and applying Bayesian optimization for device sizing. On a curated benchmark suite covering 13 circuit types, AnalogCoder-Pro successfully designed 28 circuits and consistently outperformed existing LLM-based methods in figures of merit.
format Preprint
id arxiv_https___arxiv_org_abs_2508_02518
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AnalogCoder-Pro: Unifying Analog Circuit Generation and Optimization via Multi-modal LLMs
Lai, Yao
Poddar, Souradip
Lee, Sungyoung
Chen, Guojin
Hu, Mengkang
Yu, Bei
Luo, Ping
Pan, David Z.
Machine Learning
Despite recent advances, analog front-end design still relies heavily on expert intuition and iterative simulations, which limits the potential for automation. We present AnalogCoder-Pro, a multimodal large language model (LLM) framework that integrates generative and optimization techniques. The framework features a multimodal diagnosis-and-repair feedback loop that uses simulation error messages and waveform images to autonomously correct design errors. It also builds a reusable circuit tool library by archiving successful designs as modular subcircuits, accelerating the development of complex systems. Furthermore, it enables end-to-end automation by generating circuit topologies from target specifications, extracting key parameters, and applying Bayesian optimization for device sizing. On a curated benchmark suite covering 13 circuit types, AnalogCoder-Pro successfully designed 28 circuits and consistently outperformed existing LLM-based methods in figures of merit.
title AnalogCoder-Pro: Unifying Analog Circuit Generation and Optimization via Multi-modal LLMs
topic Machine Learning
url https://arxiv.org/abs/2508.02518