AI Trends

Tibo Says GPT-6.2 Astra Excels at Challenging Chip Optimization Tasks

TiboUpdated Nov 19 09:28

AI summary

Tibo believes GPT-6.2 Astra is highly capable of handling challenging tasks and shared the team's results in optimizing AMD and Jalapeño chips. He said the team has received feedback about the model's communication style and is working to improve it; for the most difficult tasks, he currently sees the model as the leader.

Why read this

AI score 68

Drawing on chip optimization results, the author assesses GPT-6.2 Astra's ability to handle complex tasks while also responding to feedback about its communication style.

Topics

Tags

Quoted post

OpenAI

@OpenAI

Working with @AMD, our models identified up to 14% PPA improvement on flagship GPU IP in less than five months—building on designs AMD’s engineers had already spent years optimizing. The same approach helped improve Jalapeño, our first custom inference chip, with gains verified on real silicon. New evaluations also show our models getting better at certain chip design tasks. We’re sharing these results and how we’re working with @Synopsys to bring our model capabilities and their trusted EDA design and validation tools to more chip designers. https://t.co/4BNcPIqfIQ

View original

Working with @AMD, our models identified up to 14% PPA improvement on flagship GPU IP in less than five months—building on designs AMD’s engineers had already spent years optimizing. The same approach helped improve Jalapeño, our first custom inference chip, with gains verified on real silicon. New evaluations also show our models getting better at certain chip design tasks. We’re sharing these results and how we’re working with @Synopsys to bring our model capabilities and their trusted EDA design and validation tools to more chip designers. https://t.co/4BNcPIqfIQ

View original post

Body

GPT-6.2 Astra is pretty good at hard things. Here the team shared some results on AMD and Jalapeño chip optimizations.

I do get the feedback on its communication style and we're working on that, but for really the hardest tasks in the world it does seem to be in a league of its own at the moment.

Source

Tibo