VLM-CAD: VLM-Optimized Collaborative Agent Design Workflow for Analog Circuit Sizing

2026-01-12

AI Summary

This work addresses the high-dimensional, complex trade-offs inherent in analog-mixed signal circuit sizing, a task where existing methods struggle to effectively leverage circuit schematics and lack interpretability. The authors propose a multi-agent collaborative optimization framework integrating vision-language models, which employs an Image2Net module to structurally parse circuit diagrams, JSON-based semantic descriptions, an interpretable trust-region Bayesian optimizer (ExTuRBO), and dual-granularity sensitivity analysis to enable reasoning-driven sizing and collaborative warm-starting. Evaluated across technology nodes from 180 nm to 45 nm, the approach achieves 100% success in designing complementary-input and class-AB output-stage amplifiers within under 43 minutes total runtime, striking a favorable balance between power consumption and performance while significantly advancing automation and industrial interpretability.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Optimization
• Computer Vision: Learning & Optimization for CV

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Analog mixed-signal circuit sizing involves complex trade-offs within high-dimensional design spaces. Existing automatic analog circuit sizing approaches rely solely on netlists, ignoring the circuit schematic, which hinders the cognitive link between the schematic and its performance. Furthermore, the black-box nature of machine learning methods and hallucination risks in large language models fail to provide the necessary ground-truth explainability required for industrial sign-off. To address these challenges, we propose a Vision Language Model-optimized collaborative agent design workflow (VLM-CAD), which analyzes circuits, optimizes DC operating points, performs inference-based sizing, and executes external sizing optimization. We integrate Image2Net to annotate circuit schematics and generate a structured JSON description for precise interpretation by Vision Language Models. Furthermore, we propose an Explainable Trust Region Bayesian Optimization method (ExTuRBO) that employs collaborative warm-start from agent-generated seeds and offers dual-granularity sensitivity analysis for external sizing optimization, supporting a comprehensive final design report. Experiment results on amplifier sizing tasks using 180nm, 90nm, and 45nm Predictive Technology Models demonstrate that VLM-CAD effectively balances power and performance while maintaining physics-based explainability. VLM-CAD meets all specification requirements while maintaining low power consumption in optimizing an amplifier with a complementary input and a class-AB output stage, with a total runtime under 66 minutes across all experiments on two amplifiers.

Problem

• analog circuit sizing
• circuit schematics
• explainability
• high-dimensional design space
• automatic sizing

Innovation

• Vision Language Model
• Collaborative Agent
• Explainable Optimization
• Analog Circuit Sizing
• Image2Net