OpenAI says models added a year of AMD GPU optimization
Original title: OpenAI says its models delivered a year's worth of chip optimization on top of AMD's own engineering
OpenAI claims up to 14 percent PPA gain on AMD flagship GPU IP, with some changes already sent to manufacturing.
Working with AMD, OpenAI says its models improved power, performance and area on AMD flagship GPU IP by up to 14 percent in less than five months, which it equates to about a year of earlier engineering. OpenAI says AMD has folded some changes into a design sent to manufacturing. The same method aided Jalapeño, OpenAI’s first custom inference chip, and GPT-6.2 Astra led tests on open RISC-V cores. OpenAI is also building GPT-Synopsys with Synopsys.
Unconfirmed: The 14 percent gain, the year-equivalent claim and Anthropic’s three-day figure are company statements, not independent measurements.
- OpenAI says GPT-6.2 Astra ranked best across generations on open RISC-V cores.
- GPT-Synopsys is intended to use Synopsys EDA tools for AI-assisted design.
- Anthropic says Mythos cut an AMD MI355X task from months to three days.
- Anthropic says it now runs Claude on four chip families.
From the original article
1 min read at the-decoder.comWorking with AMD, OpenAI achieved up to 14 percent PPA improvement on AMD's flagship GPU IP in less than five months. According to OpenAI, that's roughly a year's worth of progress, and AMD has already folded the changes into a design sent to production. Anthropic is also applying AI to chip work to optimize cost and performance per dollar.
By Matthias Bastian · the-decoder.com
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.