MuaLLM: A Multimodal Large Language Model Agent for Circuit Design Assistance with Hybrid Contextual Retrieval-Augmented Generation

TL;DR

MuaLLM is a multimodal LLM agent designed for circuit design assistance, integrating hybrid retrieval-augmented generation, iterative reasoning, and multimodal data processing to improve efficiency, scalability, and accuracy in complex circuit research tasks.

Contribution

This paper introduces MuaLLM, a novel multimodal LLM agent with a hybrid RAG framework and ReAct workflow, enabling scalable, cost-effective, and accurate circuit design assistance beyond traditional model limits.

Findings

Achieves 90.1% recall on RAG-250 dataset.

Attains 86.8% accuracy on Reas-100 reasoning dataset.

Operates up to 10x less costly and 1.6x faster than conventional models.

Abstract

Conducting a comprehensive literature review is crucial for advancing circuit design methodologies. However, the rapid influx of state-of-the-art research, inconsistent data representation, and the complexity of optimizing circuit design objectives make this task significantly challenging. In this paper, we propose MuaLLM, an open-source multimodal Large Language Model (LLM) agent for circuit design assistance that integrates a hybrid Retrieval-Augmented Generation (RAG) framework with an adaptive vector database of circuit design research papers. Unlike conventional LLMs, the MuaLLM agent employs a Reason + Act (ReAct) workflow for iterative reasoning, goal-setting, and multi-step information retrieval. It functions as a question-answering design assistant, capable of interpreting complex queries and providing reasoned responses grounded in circuit literature. Its multimodal capabilities enable processing of both textual and visual data, facilitating more efficient and comprehensive analysis. The system dynamically adapts using intelligent search tools, automated document retrieval from the internet, and real-time database updates. Unlike conventional approaches constrained by model context limits, MuaLLM decouples retrieval from inference, enabling scalable reasoning over arbitrarily large corpora. At the maximum context length supported by standard LLMs, MuaLLM remains up to 10x less costly and 1.6x faster while maintaining the same accuracy. This allows rapid, no-human-in-the-loop database generation, overcoming the bottleneck of simulation-based dataset creation for circuits. To evaluate MuaLLM, we introduce two custom datasets: RAG-250, targeting retrieval and citation performance, and Reasoning-100 (Reas-100), focused on multistep reasoning in circuit design. MuaLLM achieves 90.1% recall on RAG-250, and 86.8% accuracy on Reas-100.