Design Conductor 2.0: An agent builds a TurboQuant inference accelerator in 80 hours

2026-05-06 · The Verkor Team, Ravi Krishna, Suresh Krishna, David Chin arxiv

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.

📄 PDF Abstract BibTeX arXiv:2605.05170

Code (0)

등록된 구현이 없습니다.

Design Conductor: An agent autonomously builds a 1.5 GHz Linux-capable RISC-V CPU

2026-02-06 · The Verkor Team, Ravi Krishna, Suresh Krishna, David Chin arxiv

Design Conductor (DC) is an autonomous agent which applies the capabilities of frontier models to build semiconductors end-to-end -- that is, from concept to verified, tape-out ready GDSII (layout CAD file). In 12 hours …

A Note on TurboQuant and the Earlier DRIVE/EDEN Line of Work

2026-04-20 · Ran Ben-Basat, Yaniv Ben-Itzhak, Gal Mendelson, Michael Mitzenmacher 외 arxiv

This note clarifies the relationship between the recent TurboQuant work and the earlier DRIVE (NeurIPS 2021) and EDEN (ICML 2022) schemes. DRIVE is a 1-bit quantizer that EDEN extended to any $b>0$ bits per coordinate; w…

Revisiting RaBitQ and TurboQuant: A Symmetric Comparison of Methods, Theory, and Experiments

2026-04-21 · Jianyang Gao, Yutong Gou, Yuexuan Xu, Jifan Shi 외 arxiv

This technical note revisits the relationship between RaBitQ and TurboQuant under a unified comparison framework. We compare the two methods in terms of methodology, theoretical guarantees, and empirical performance, usi…

Learning to Orchestrate Agents in Natural Language with the Conductor

2025-12-04 · Stefan Nielsen, Edoardo Cetin, Peter Schwendeman, Qi Sun 외 arxiv

Powerful large language models (LLMs) from different providers have been expensively trained and finetuned to specialize across varying domains. In this work, we introduce a new kind of Conductor model trained with reinf…

Reinforcement Learning

Block-Sphere Vector Quantization

2026-05-19 · Heesang Ann, Joongkyu Lee, Min-hwan Oh arxiv

Vector quantization is a fundamental primitive for scalable machine learning systems, enabling memory-efficient storage, fast retrieval, and compressed inference. Recent rotation-based quantizers such as EDEN, RabitQ, an…