OpenAI models deliver up to 14% PPA gains on AMD GPU IP in 5 months, helping tape out Jalapeño
gdb · x · 2026-11-19
- OpenAI and AMD report that OpenAI's models identified up to 14% PPA (power/performance/area) improvements on flagship GPU IP in under five months, building on designs AMD engineers had already optimized for years.
- Greg Brockman says this equals roughly an extra year of progress at the frontier of chip design, with some results included in a recently taped-out design.
- The same approach improved Jalapeño, OpenAI's first custom inference chip, with gains verified on real silicon.
- OpenAI is also partnering with Synopsys to bring model capabilities into its trusted EDA design and validation tools.
Related event: OpenAI Models Deliver Up to 14% PPA Gains on AMD Flagship GPU IP(6 posts)→
More from Infra
- NVIDIA's Qwen3.5-4B draft model for Qwen3.8-27B trends on Hugging Face — nvidia · 2026-11-19
- Banks offload $800M of AI data center debt, including loans tied to Oracle's Stargate — SumitGup · 2026-11-19
- BIS interim rule targets cloud remote access to US chips; CNAS unpacks what it gets right and wrong — paul_scharre · 2026-11-19
- Japanese neocloud Datasection slashes revenue guidance 89%, stock limit-down — dylan522p · 2026-11-19
- Matt Wolfe's video essay: the reason no one can catch NVIDIA — Matt Wolfe · 2026-11-19
- Developers push back against the thin-client future: 'ram altman doesn't want you to have RAM' — sull · 2026-11-19