# Agentic AI Explores Heterogeneous Chip and Interconnect Architectures (Argonne, UIUC)

DevFeed: [Agentic AI Explores Heterogeneous Chip and Interconnect Architectures (Argonne, UIUC)](<https://devfeed.tech/articles/agentic-ai-explores-heterogeneous-chip-and-interconnect-architectures-argonne-uiuc-90103.md>)

Original publisher: [Read original article](<https://semiengineering.com/agentic-ai-explores-heterogeneous-chip-and-interconnect-architectures-argonne-uiuc/>)

Author: Technical Paper Link

Published: 2026-10-19T16:02:46Z

Content type: article

Language: en

Sources: [Semiconductor Engineering](<https://devfeed.tech/sources/semiconductor-engineering.md>)

Topics: [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Chip design](<https://devfeed.tech/topics/chip-design.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>), [InfiniBand](<https://devfeed.tech/topics/infiniband.md>), [Legacy Modernization](<https://devfeed.tech/topics/legacy-modernization.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-ml-dl](<https://devfeed.tech/tags/ai-ml-dl.md>), [archagent](<https://devfeed.tech/tags/archagent.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [argonne-national-laboratory](<https://devfeed.tech/tags/argonne-national-laboratory.md>), [chip](<https://devfeed.tech/tags/chip.md>), [co-design](<https://devfeed.tech/tags/co-design.md>), [communication-data-movement](<https://devfeed.tech/tags/communication-data-movement.md>), [design-space-exploration](<https://devfeed.tech/tags/design-space-exploration.md>), [design-verification](<https://devfeed.tech/tags/design-verification.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [hbm](<https://devfeed.tech/tags/hbm.md>), [heterogeneous-systems](<https://devfeed.tech/tags/heterogeneous-systems.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [interconnects](<https://devfeed.tech/tags/interconnects.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-agents](<https://devfeed.tech/tags/llm-agents.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [power-performance](<https://devfeed.tech/tags/power-performance.md>), [system-design](<https://devfeed.tech/tags/system-design.md>), [technical-papers](<https://devfeed.tech/tags/technical-papers.md>), [university-of-illinois-urbana-champaign](<https://devfeed.tech/tags/university-of-illinois-urbana-champaign.md>)

## AI overview

Researchers at Argonne National Laboratory and the University of Illinois Urbana-Champaign describe ArchAgent, a hierarchical framework that uses an LLM-based planner and specialist agents to explore heterogeneous AI and HPC system designs across chip, scale-up, and scale-out domains. The paper addresses a design space spanning compute architectures, memory technologies, and interconnects, with different configurations suited to LLM training, serving, and AI-coupled HPC workloads.

## Source excerpt

Researchers at Argonne National Laboratory and the University of Illinois Urbana-Champaign published a technical paper titled "Agentic AI for Heterogeneous System Design Across Chip, Scale-Up, and Scale-Out Domains." Abstract Excerpt: "The design space of modern AI/HPC systems has expanded rapidly across compute architectures, including GPUs, dataflow processors, and RISC-V accelerators; memory technologies, including HBM, GDDR,... " read more The post Agentic AI Explores Heterogeneous Chip and Interconnect Architectures (Argonne, UIUC) appeared first on Semiconductor Engineering.