How AI Created the World’s First Fully Autonomous CPU – Qǐ Méng 1
Chinese researchers used generative AI to design Qǐ Méng 1, the first completely self‑generated CPU chip that runs Linux and rivals early Intel 486 performance, showcasing the rapid rise of AI‑driven semiconductor design.
According to reports, the Institute of Computing Technology of the Chinese Academy of Sciences used AI technology to design the world’s first fully autonomous, automatically generated CPU chip – Qǐ Méng 1.
Generative AI is demonstrating huge potential, and the semiconductor field has long researched AI‑automated chip design.
Recently, the institute achieved a global first by fully automating chip design with AI, naming the processor Qǐ Méng 1, which can run Linux and delivers performance comparable to Intel’s 486.
Qǐ Méng 1 is based on the RISC‑V instruction set, features a 32‑bit architecture, and completed tape‑out verification in December 2021.
Subsequently, the chip successfully booted a Linux operating system and ran the SPEC CPU2000 benchmark, with results judged to be on par with Intel’s 40486.
This is the world’s first fully autonomous CPU chip, fabricated with a 65 nm process, operating at 300 MHz, and its related research paper has been published.
Before Qǐ Méng 1, there were already AI‑automated chip design studies; after the rise of ChatGPT, attempts increased, but those chips were generally simple and of limited practical significance.
Qǐ Méng 1 was designed by researchers using the BSD (Binary Speculation Diagram) algorithm, generating 4 million logic gates in five hours—a scale 4 000 times larger than chips designed by GPT‑4.
Although the scale of Qǐ Méng 1’s CPU does not yet match top‑tier CPUs, AI technology is advancing rapidly, and AI‑designed chips are expected to become mainstream, dramatically improving design efficiency.
Researchers at NYU Tandon School of Engineering have already used simple English dialogue with ChatGPT‑4 to generate a micro‑processor chip, a breakthrough that can accelerate chip development and potentially enable non‑experts to design chips.
While Qǐ Méng 1’s scale cannot yet compete with current flagship CPUs, future cross‑disciplinary AI approaches may allow AI‑generated CPUs to reach or surpass human‑expert designs within five to ten years.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
21CTO
21CTO (21CTO.com) offers developers community, training, and services, making it your go‑to learning and service platform.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
