OpenAI Unveils Custom AI Inference Chip 'Jalapeño' with Broadcom
OpenAI and Broadcom unveiled 'Jalapeño,' a custom AI inference chip designed for OpenAI's systems.
"OpenAI just dropped its own AI chip, Jalapeño, with Broadcom. They're going all-in on custom hardware to make AI faster and cheaper. Watch out, chip giants!"
OpenAI has revealed its first custom-built inference processor, named Jalapeño, developed in collaboration with Broadcom. The new processor is specifically designed for the unique needs of OpenAI's inference systems, according to announcements from OpenAI and Broadcom. The chip, an ASIC, is intended to power current and future large language models.
This custom chip development marks OpenAI's entry into the custom-chip race, aiming to boost its AI infrastructure. The partnership was officially announced on Wednesday, with the chip designed to optimize the process of running pre-built AI models in response to user commands. OpenAI stated that its own AI models assisted in the chip's development.
The introduction of Jalapeño highlights OpenAI's strategy to optimize its infrastructure across the stack, from chip architecture to deployment systems. By developing purpose-built chips, OpenAI aims to make its models faster, more reliable, and more affordable for users, particularly by reducing inference costs.
This development signifies a strategic move by OpenAI to gain greater control over its AI infrastructure and potentially reduce operational costs. For businesses leveraging AI, this could lead to more efficient and cost-effective AI services in the future, impacting the economics of AI across various applications.
Relevant tools
Find the right AI tool for your business
Chat with Insta and get matched to the right tool in seconds.
Try Insta Tool Finder →