Today, most AI systems rely on powerful GPUs. While highly effective, these processors were originally designed for broad graphics processing, making them incredibly expensive and power-hungry when handling massive, everyday consumer traffic.<\/p>\n\n\n\n
OpenAI\u2019s new chip shifts the strategy toward total architectural optimization.<\/p>\n\n\n\n
“Jalape\u00f1o was designed from the ground up for LLM inference using detailed insights from our close collaboration with OpenAI researchers,\u201d Richard Ho, Head of OpenAI\u2019s Hardware Program said at the launch.<\/p>\n\n\n\n
\u201cWe optimized the architecture around the kernels, memory movement, networking, and serving patterns that matter most for frontier AI models. Based on early testing, Jalape\u00f1o will efficiently execute our most important workloads close to the hardware’s theoretical limits.”<\/p>\n\n\n\n
What changes with Jalape\u00f1o?<\/strong><\/p>\n\n\n\n By tailoring the hardware directly to the software, OpenAI and its partners say the chip delivers a few massive upgrades.<\/p>\n\n\n\n First, it brings drastically lower costs, with early industry reports indicating it could cut inference costs by roughly 50% compared to standard cloud infrastructure.<\/p>\n\n\n\n Second, the chip offers massive energy efficiency by delivering substantially better performance-per-watt than current state-of-the-art options. This allows data centers to pull more compute capacity out of the exact same power grid allocation.<\/p>\n\n\n\n Finally, the project achieved unprecedented development speed, moving from an initial blank-slate concept to final manufacturing readiness (tape-out) in just nine months. This process was partially accelerated by OpenAI using its own AI models to optimize the layout and architecture.<\/p>\n\n\n\n OpenAI co-developed the chip with semiconductor giant Broadcom, which provided critical silicon implementation and networking technology, including its Tomahawk networking platform. Physical server hardware, circuit boards, and rack integration are being handled by engineering partner Celestica, with manufacturing slated for Taiwan Semiconductor Manufacturing Company (TSMC).<\/p>\n\n\n\n OpenAI, the company behind ChatGPT, has announced a new custom-built computer chip called \u201cJalape\u00f1o.\u201d It is designed to make artificial intelligence faster and significantly cheaper to run at scale.<\/p>\n","protected":false},"author":139,"featured_media":2000117097,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[9],"tags":[151],"byline":[201],"hashtag":[],"class_list":["post-2000117095","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-science-technology","tag-editors-choice","byline-rania-umutoni"],"bylines":[{"id":201,"name":"Rania Umutoni","slug":"rania-umutoni","description":"","image":{"id":0,"url":"https:\/\/secure.gravatar.com\/avatar\/?s=96&d=mm&f=y&r=g","alt":"Default avatar","title":"Default avatar","caption":"","mime_type":"image\/jpeg","sizes":[]},"user_id":139}],"contributors":[{"id":201,"name":"Rania Umutoni","slug":"rania-umutoni","description":"","image":{"id":0,"url":"https:\/\/secure.gravatar.com\/avatar\/?s=96&d=mm&f=y&r=g","alt":"Default avatar","title":"Default avatar","caption":"","mime_type":"image\/jpeg","sizes":[]},"user_id":139}],"featured_image":{"id":2000117097,"url":"https:\/\/cdn.igihe.com\/en\/2026\/06\/hero-1782359752749.png","alt":"","caption":"","mime_type":"image\/png","width":1536,"height":1024,"sizes":{"thumbnail":{"url":"https:\/\/cdn.igihe.com\/en\/2026\/06\/hero-1782359752749.png","width":1536,"height":1024},"medium":{"url":"https:\/\/cdn.igihe.com\/en\/2026\/06\/hero-1782359752749.png","width":1536,"height":1024},"medium_large":{"url":"https:\/\/cdn.igihe.com\/en\/2026\/06\/hero-1782359752749.png","width":1536,"height":1024},"large":{"url":"https:\/\/cdn.igihe.com\/en\/2026\/06\/hero-1782359752749.png","width":1536,"height":1024},"full":{"url":"https:\/\/cdn.igihe.com\/en\/2026\/06\/hero-1782359752749.png","width":1536,"height":1024}}},"_links":{"self":[{"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/posts\/2000117095","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/users\/139"}],"replies":[{"embeddable":true,"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/comments?post=2000117095"}],"version-history":[{"count":3,"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/posts\/2000117095\/revisions"}],"predecessor-version":[{"id":2000117192,"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/posts\/2000117095\/revisions\/2000117192"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/media\/2000117097"}],"wp:attachment":[{"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/media?parent=2000117095"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/categories?post=2000117095"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/tags?post=2000117095"},{"taxonomy":"byline","embeddable":true,"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/byline?post=2000117095"},{"taxonomy":"hashtag","embeddable":true,"href":"https:\/\/new.igihe.com\/english\/wp-json\/wp\/v2\/hashtag?post=2000117095"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"\"](\"https:\/\/cdn.igihe.com\/en\/2026\/06\/hero-1782359752749.png\")