Design Conductor 2.0: An agent builds a TurboQuant inference accelerator in 80 hours
The Verkor Team, Ravi Krishna, Suresh Krishna, David Chin
TLDR
Design Conductor 2.0, an advanced LLM agent, autonomously builds complex hardware designs, including a TurboQuant inference accelerator, in just 80 hours.
Key contributions
- Introduces Design Conductor 2.0, a multi-agent LLM system handling 80x larger tasks autonomously.
- Autonomously designed "VerTQ," a TurboQuant LLM inference accelerator from an arXiv paper.
- VerTQ features 5129 FP16/32 units and a 240-cycle pipeline, mapped to FPGA at 125 MHz.
- Achieved 5.7 mm^2 in TSMC 16FF for 8 attention pipes, demonstrating efficient hardware synthesis.
Why it matters
This paper demonstrates a significant leap in LLM agent capabilities for autonomous hardware design. It showcases the potential for agents to rapidly synthesize complex, high-performance accelerators directly from high-level specifications, accelerating hardware development.
Original Abstract
Driven by a rapid co-evolution of both harness and underlying models, LLM agents are improving at a dizzying pace. In our prior work (performed in Dec. 2025), we introduced "Design Conductor" (or just "Conductor"), a system capable of building a 5-stage Linux-capable RISC-V CPU in 12 hours. In this work, we introduce an updated multi-agent harness powered by frontier models released in April 2026, which is able to handle 80x larger tasks, at higher quality, fully autonomously. Following a brief introduction, we examine 4 designs that the system produced autonomously, including "VerTQ", an LLM inference accelerator which hard-wires support for TurboQuant in a 240-cycle pipeline, starting from the TurboQuant arXiv paper. VerTQ includes heavy compute processing, with 5129 FP16/32 units; the design was mapped to an FPGA at 125 MHz and consumes 5.7 mm^2 in TSMC 16FF (8 attention pipes). We review the key new characteristics that enabled these results. Finally, we analyze Design Conductor's token usage and other empirical characteristics, including its limitations.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.