Skip to content
NewA calmer, faster ZeroHour

OpenAI models used to improve Jalapeño and AMD chip designs

Original title: Advancing chip design with OpenAI models

OpenAI News ()7 min read1
Part of a story covered by 2 sourcesOpenAI says models cut Jalapeño area and improved AMD GPU IPMerged summary & timeline
AI summary · x-ai/grok-4.7Summaries by Vani AI

OpenAI reports about 10–15% cell-area cuts on Jalapeño blocks and up to 14% PPA gains on AMD GPU IP.

OpenAI says successive models, starting with GPT-4o in early 2025, proposed power, performance and area changes to its Jalapeño inference chip that were silicon-proven at bring-up. With AMD, the company reports up to 14% improvement versus optimized GPU IP baselines and single-digit PPA gains on an upcoming flagship CPU; some changes are already in a manufactured design. OpenAI also scored model generations on PicoRV32 and Ibex and is working with Synopsys to connect models to existing EDA tools.

Unconfirmed: Area, latency and PPA figures, and the claim of roughly a year of progress in under five months, are OpenAI’s estimates.

Key facts
  • GPT-4o began assisting Jalapeño design in early 2025.
  • One multiplier redesign cut cell area about 15% and latency from two cycles to one.
  • AMD included some model-written changes in a design already sent for manufacturing.
  • GPT-6.2 Astra had the highest average score on PicoRV32 and Ibex at about 112,000 tokens.
  • OpenAI is working with Synopsys to use models inside existing EDA tools.
Explain this AI news with Vani AIWhat was announced, what is new and why it matters, in a new chat

From the original article

7 min read at openai.com

OpenAI models optimized circuits used in Jalapeño silicon and improved RISC‑V core designs, showing progress in AI-assisted chip design with GPT‑6 Astra.

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.