In a nutshell: OpenAI unveiled its first proprietary AI chip called Jalapeno, specifically optimized for its own model requirements.
OpenAI introduced its first proprietary chip optimized for the company’s own AI models, named Jalapeno. Development took place in collaboration with external partners and was tailored directly to OpenAI’s model requirements.
OpenAI presented Jalapeno as a chip specifically developed for the requirements of the company’s proprietary AI models. The hardware was created in collaboration with OpenAI employees and involved service providers from the semiconductor industry.
The move follows the trend among leading AI companies to further verticalize their infrastructure. By developing hardware specifically tailored to the architecture and requirements of proprietary models, efficiency and performance in training and operation can be optimized.
Details on specifications, manufacturing process, performance characteristics, or planned deployment timeline were not mentioned in the announcement. OpenAI merely confirmed the completion and presentation of the chip.
Source: www.golem.de · Published 25 June 2026
Lumi AI News — AI-assisted curation in accordance with Article 50 EU AI Act. Paraphrase and classification by Lumi News Pipeline v1.7.1.