AutoEDA Introduces Natural-Language RTL-to-GDSII Automation System
Researchers Yiyi Lu et al. (v2 Feb 24, 2026) present AutoEDA, a framework using the Model Context Protocol (MCP) to enable end-to-end natural language control of RTL-to-GDSII design flows. AutoEDA deploys MCP-based servers for task decomposition, tool selection, automated error handling, locally fine-tuned LLM agents, a benchmark generation pipeline, and Tcl-specific CodeBLEU enhancements. Experiments show up to 9.9× higher accuracy versus naive methods and roughly 97% token usage reduction.
Key Points
- 1Demonstrates AutoEDA enabling end-to-end natural-language control of RTL-to-GDSII flows using MCP servers
- 2Reduces reliance on external APIs and increases confidentiality via locally fine-tuned LLM agents
- 3Improves automation accuracy up to 9.9× and cuts token usage by ~97%, speeding EDA iteration cycles
Scoring Rationale
Strong practical framework and sizable empirical gains plus domain protocol, limited by single preprint source and niche EDA scope.
Sources
Public references used for this report.
Practice with real Logistics & Shipping data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Logistics & Shipping problems