OpenAI details custom Jalapeño AI inference ASIC and its AI-assisted design
OpenAI has revealed details about its custom AI inference ASIC, named Jalapeño, which was designed with the assistance of AI. The company's head of hardware confirmed that the Jalapeño ASICs are intended for OpenAI's internal use and are being deployed alongside AMD EPYC 'Turin' CPUs, specifically not Nvidia's Vera. While primarily for internal operations, OpenAI has indicated that a broader rollout of the custom chip could be a future possibility. The information regarding the ASIC's design, purpose, and deployment comes directly from OpenAI. This marks a significant step in OpenAI's internal compute infrastructure.
5 articles from 2 outlets covered this story. The underlying claim is sourced from a shipped.
What do all outlets agree on?
2 outlets covered “OpenAI details custom Jalapeño AI inference ASIC and its AI-assisted design”. All of them report the following:
- OpenAI has a custom AI inference ASIC
- The ASIC is named Jalapeño
- Jalapeño was designed with AI assistance
- Jalapeño is for OpenAI's internal use
- Jalapeño is deployed with AMD EPYC 'Turin' CPUs
- OpenAI's head of hardware discussed the ASIC
Which outlets covered this?
All 5 articles found on this story, grouped by the stance of the piece. Every link goes to the original publisher.