OpenAI team member Tibo evaluates GPT-6.2 Astra and cites AMD and Jalapeño chip optimization results
Original title:GPT-6.2 Astra is pretty good at hard things. Here the team shared some results on AMD and Jalapeño chip optimizations.
AI summary
OpenAI member Tibo cited team results and said GPT-6.2 Astra excels at extremely difficult tasks, while acknowledging feedback on his communication style and saying he is working to improve. The cited content says a collaboration with AMD raised the PPA of flagship GPU IP by up to 14% in under five months, that the same approach improved OpenAI's first in-house reasoning chip Jalapeño with validation on real silicon, and that a collaboration with Synopsys brought model capabilities into its EDA design and verification tools. Details at https://openai.com/index/advancing-chip-design-with-our-models/
Key Takeaways
- OpenAI member Tibo cited team results and said GPT-6.
- 2 Astra excels at extremely difficult tasks,while acknowledging feedback on his communication…
- The cited content says a collaboration with AMD raised the PPA of flagship GPU IP by up to 14% in…
- Details at
Quick Facts
- Published
- Source
- X:Tibo (@thsottiaux)
- Category
- Model Releases
- AI score
- 63
作者以内部视角评价 GPT-6.2 Astra 在最难任务上的表现,并引 OpenAI 与 AMD 合作的芯片优化结果作佐证。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: OpenAI team member Tibo evaluates GPT-6.2 Astra and cites AMD and Jalapeño chip optimization results. November 19, 2026. https://new.nat6.net/en/news/bnqsw7ispucfily4bdq9vz33j/