CHICO-Agent: An LLM Agent for the Cross-layer Optimization of 2.5D and 3D Chiplet-based Systems

TL;DR

CHICO-Agent is an LLM-based framework that optimizes complex 2.5D/3D chiplet systems across multiple design trade-offs, outperforming traditional methods.

Contribution

It introduces a novel LLM-driven multi-agent framework with a persistent knowledge base for efficient system-level chiplet design optimization.

Findings

CHICO-Agent finds lower-cost configurations than simulated-annealing.

Provides an interpretable audit trail for design decisions.

Effectively manages the complex design space of 2.5D/3D chiplet systems.

Abstract

The rapid growth of large language models (LLMs) and AI workloads has pushed monolithic silicon to its reticle and economic limits, accelerating the adoption of 2.5D/3D chiplet systems. However, these systems increase design complexity by requiring co-design across multiple levels of the computing stack, including application, architecture, chip, and package. The resulting design space is highly combinatorial, with trade-offs among latency, energy, area, and cost. To address this challenge, we propose CHICO-Agent, an LLM-driven optimization framework for 2.5D/3D chiplet-based systems. CHICO-Agent maintains a persistent knowledge base to capture parameter-outcome trends and coordinates exploration through an admin-field multi-agent workflow. Compared with a simulated-annealing baseline, CHICO-Agent finds lower-cost configurations and provides an interpretable audit trail for designers.