OpenAI Report Says LLMs Cut AMD Chip Optimization Cycle to Under Five Months
An OpenAI technical report says its large language models helped design AMD chips, compressing roughly a year of engineering optimization into less than five months and achieving PPA improvements of up to 14% in flagship GPU circuit designs. After the models rewrote portions of the SystemVerilog RTL code, the changes were verified using Synopsys EDA tools, and some optimizations have been incorporated into designs recently sent for fabrication. The 14% figure represents combined gains in power, performance, and area, and does not mean that actual GPU performance directly improved by 14%.
Why read this
SelectedAI score 77
The report details how large language models optimize mature chip designs and deliver PPA gains, while clarifying the distinction between metric improvements and actual product performance.
Topics
Body · AI translation
AI Chip Design Really Works! A Year’s Work Finished in Five Months: AMD Has Already Sent Designs to the Fab for Tape-Out Kuai Technology Client Download — Exclusive Early Look November 21, 2026, 13:20:45, from Henan
Kuai Technology reported on November 21 that OpenAI released its latest technical report on November 18, disclosing a rather interesting development: its large models have become deeply involved in AMD’s chip design process, achieving PPA improvements of up to 14% in the circuit design of the chipmaker’s flagship GPU.
Moreover, some AI-generated optimizations have already been incorporated into chip designs recently sent to the fab for manufacturing. According to OpenAI, the collaboration compressed roughly a year of engineering optimization work into less than five months.
Unlike the common use of AI to write software code, the optimization target in this case was hardware description language. OpenAI’s models directly rewrote portions of AMD’s SystemVerilog RTL code.
This is the layer that defines how circuits perform computation, control, and data transmission. After the changes, Synopsys EDA tools were used for functional verification to ensure that the circuits’ behavior remained unchanged.
What is particularly noteworthy is that these optimizations were extracted from designs that AMD engineers had already refined over many years. Even against highly optimized benchmarks, the models still uncovered room for PPA improvements of up to 14%.
As for implementation progress, AMD has already integrated some of the changes into chip designs recently sent to the fab for manufacturing. The remaining optimizations have also entered production design files for future products. Their scope extends beyond GPUs—AMD has also used model-generated designs to verify single-digit percentage PPA improvements in key modules of its next-generation flagship CPU.
It should be emphasized, however, that the 14% figure refers to gains in PPA and does not mean that actual GPU performance increased directly by 14%. OpenAI did not break down the respective contributions from power consumption, performance, and area, while AMD has not confirmed which specific products will use these optimizations.
Given that RDNA5, CDNA5, and Zen6 are all next-generation architectures that AMD is preparing to launch, outside observers will inevitably associate the optimizations with them, but there has been no official statement to that effect.
The bigger highlight of the entire development is the closed loop: AI models are now helping design the hardware used to run those same models. How much practical benefit these optimizations will ultimately bring to AMD’s commercial GPUs and CPUs will depend on the performance of subsequent products.
“Special Notice: The content of the above work (including any videos, images, or audio) was uploaded and published by users of Dafeng Hao, a self-media platform under Phoenix New Media. This platform merely provides information storage services.
Notice: The content above (including the videos, pictures and audios if any) is uploaded and posted by the user of Dafeng Hao, which is a social media platform and merely provides information storage space services.”
AI Trends