AnalogCoder-Pro: Unifying Analog Circuit Generation and Optimization via Multi-modal LLMs

2025/08/04 by Lai, Yao, Poddar, Souradip, Lee, Sungyoung +5 · 3 citations
#FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2508.02518

Abstract

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.

Citations

• Dataset Construction for Training LLM to Learn Analog Circuit Knowledge
• MuaLLM: A Multimodal Large Language Model Agent for Circuit Design Assistance with Hybrid Contextual Retrieval-Augmented Generation
• MMCircuitEval: A Comprehensive Multimodal Circuit-Focused Benchmark for Evaluating LLMs
• Evaluating LLM-based Workflows for Switched-Mode Power Supply Design
• LaMAGIC2: Advanced Circuit Formulations for Language Model-Based Analog Topology Generation
• ScaleRTL: Scaling LLMs with Reasoning Data and Test-Time Compute for Accurate RTL Code Generation
• AUTOCIRCUIT-RL: Reinforcement Learning-Driven LLM for Automated Circuit Topology Generation
• QiMeng-CodeV-R1: Reasoning-Enhanced Verilog Generation
• AMSbench: A Comprehensive Benchmark for Evaluating MLLM Capabilities in AMS Circuits
• RTL++: Graph-enhanced LLM for RTL Code Generation
• CircuitFusion: Multimodal Circuit Representation Learning for Agile Chip Design
• A Survey of Circuit Foundation Model: Foundation AI Models for VLSI Circuit Design and EDA
• AnalogGenie: A Generative Engine for Automatic Discovery of Analog Circuit Topologies
• Divergent Thoughts toward One Goal: LLM-based Multi-Agent Collaboration System for Electronic Design Automation
• A Survey of Research in Large Language Models for Electronic Design Automation
• AnalogXpert: Automating Analog Topology Synthesis by Incorporating Circuit Design Expertise into Large Language Models
• Masala-CHAI: A Large-Scale SPICE Netlist Dataset for Analog Circuits by Harnessing AI
• LEDRO: LLM-Enhanced Design Space Reduction and Optimization for Analog Circuits
• SPICEPilot: Navigating SPICE Code Generation and Simulation with AI Guidance
• AmpAgent: An LLM-based Multi-Agent System for Multi-stage Amplifier Schematic Design from Literature for Process and Performance Porting
• CraftRTL: High-quality Synthetic Data Generation for Verilog Code Models with Correct-by-Construction Non-Textual Representations and Targeted Code Repair
• Customized Retrieval Augmented Generation and Benchmarking for EDA Tool Documentation QA
• CodeV: Empowering LLMs with HDL Generation through Multi-Level Summarization
• Natural language is not enough: Benchmarking multi-modal generative AI for Verilog generation
• INSIGHT: Universal Neural Simulator for Analog Circuits Harnessing Autoregressive Transformers
• LLM-Enhanced Bayesian Optimization for Efficient Analog Layout Constraint Generation
• AnalogCoder: Analog Circuit Design via Training-Free Code Generation
• HDLdebugger: Streamlining HDL debugging with Large Language Models
• BetterV: Controlled Verilog Generation with Discriminative Guidance
• DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence
• LLM4EDA: Emerging Progress in Large Language Models for Electronic Design Automation
• RTLFixer: Automatically Fixing RTL Syntax Errors with Large Language Models
• AutoChip: Automating HDL Generation Using LLM Feedback
• ChipNeMo: Domain-Adapted LLMs for Chip Design
• GPT4AIGChip: Towards Next-Generation AI Accelerator Design Automation via Large Language Models
• CktGNN: Circuit Graph Neural Network for Electronic Design Automation
• Counterfactually Auditable Lifecycle Certification for Autonomous Agents
• VeriGen: A Large Language Model for Verilog Code Generation
• Voyager: An Open-Ended Embodied Agent with Large Language Models
• ChipGPT: How far are we from natural language hardware design
• Chip-Chat: Challenges and Opportunities in Conversational Hardware Design
• CodeGeeX: A Pre-Trained Model for Code Generation with Multilingual Benchmarking on HumanEval-X
• BNAI, NO-TOKEN, and MIND-UNITY: Pillars of a Systemic Revolution in Artificial Intelligence
• DNN-Opt: An RL Inspired Optimization for Analog Circuit Sizing using\n Deep Neural Networks
• LoRA: Low-Rank Adaptation of Large Language Models
• AutoCkt: Deep Reinforcement Learning of Analog Circuit Designs
• Optuna: A Next-generation Hyperparameter Optimization Framework
• AnalogFed: Federated Discovery of Analog Circuit Topologies with Generative AI

Cited by

• HeaRT: A Hierarchical Circuit Reasoning Tree-Based Agentic Framework for AMS Design Optimization
• Combining Large Language Models and Gradient-Free Optimization for Automatic Control Policy Synthesis
• Dataset Construction for Training LLM to Learn Analog Circuit Knowledge

Related

• AnalogCoder: Analog Circuit Design via Training-Free Code Generation
• AnalogTester: A Large Language Model-Based Framework for Automatic Testbench Generation in Analog Circuit Design
• LLMs for Analog Circuit Design Continuum (ACDC)
• LEDRO: LLM-Enhanced Design Space Reduction and Optimization for Analog Circuits
• MMCircuitEval: A Comprehensive Multimodal Circuit-Focused Benchmark for Evaluating LLMs