Openai and Broadcom Launch Jalapeño Chip to Enhance AI Performance
WHO:
CNN - OpenAI and Broadcom announced the development of the Jalapeño chip.
VentureBeat - Greg Brockman is the co-founder and president of OpenAI.
WHAT:
CNN - OpenAI announced that the Jalapeño chip is designed to handle workloads for future AI developments.
CNN - OpenAI is gearing up for an initial public offering that could potentially value it at a trillion dollars.
VentureBeat - OpenAI's Jalapeño chip is an Application-Specific Integrated Circuit (ASIC) designed specifically for modern large language model serving.
VentureBeat - OpenAI's partnership with Broadcom aims to deploy gigawatt-scale data centers with Microsoft and other partners beginning in 2026.
VentureBeat - OpenAI's total operational expenses for 2025 were $34 billion, resulting in an operating loss of nearly $20.92 billion.
VentureBeat - OpenAI's research and development costs in 2025 accounted for $19.18 billion, approximately 56 percent of the company's total spending.
VentureBeat - OpenAI plans to consume approximately two gigawatts of AWS's proprietary Trainium computing capacity over the next eight years as part of its agreement with Amazon.
VentureBeat - Qualcomm unveiled the Dragonfly C1000, a new data center CPU built for agentic AI, which Meta will use when production starts in 2028.
CNN - OpenAI claims that early testing shows the Jalapeño chip will deliver performance per watt substantially better than current state-of-the-art.
VentureBeat - The Jalapeño chip's engineering timeline moved from early schematics to fabrication readiness within a brief nine-month window.
VentureBeat - OpenAI has begun testing running at least one of its prior generation models, GPT‑5.3‑Codex‑Spark, on the Jalapeño chips in a test environment.
VentureBeat - The ultimate goal of the OpenAI and Broadcom partnership involves deploying gigawatt-scale data centers with Microsoft and other partners beginning in 2026.
CNN - The Jalapeño chip is designed to enhance the performance of ChatGPT and OpenAI’s coding agent, Codex.
TechRadar - The Jalapeño chip is positioned as a purpose-built processor for large language model (LLM) inference.
VentureBeat - OpenAI plans to begin rolling out the Jalapeño processors across active data centers by the end of this year.
VentureBeat - OpenAI generated $13.07 billion in revenue throughout 2025 but had total operational expenses of $34 billion, resulting in an operating loss of nearly $20.92 billion.
VentureBeat - OpenAI paid Microsoft over $10.59 billion for R&D and compute infrastructure in 2025.
WHERE:
VentureBeat - The Jalapeño chip will be deployed across active data centers.
WHEN:
VentureBeat - OpenAI and Broadcom's partnership was publicly announced in October 2025.
VentureBeat - OpenAI plans to roll out the Jalapeño processors by the end of this year.
VentureBeat - Nvidia finalized a $30 billion investment into OpenAI in February 2026.
VentureBeat - Amazon invested $50 billion into OpenAI as part of the same February 2026 funding round.
WHY:
CNN - OpenAI is positioning the new chip as a means to make its AI models more affordable and accessible to consumers.
CNN - OpenAI aims to reduce reliance on existing chip companies like Nvidia and make its AI models more affordable.
VentureBeat - OpenAI's move into chip design is intended to improve its unit economics and reduce costs associated with AI inference.
HOW:
VentureBeat - Broadcom is contributing core silicon implementation and networking technology, including Tomahawk networking silicon, to the Jalapeño chip project.
CNN - The Jalapeño chip was developed through a partnership with Broadcom and is designed specifically for modern large language models.
VentureBeat - The Jalapeño chip's engineering timeline was accelerated through a deep software-hardware co-development process using OpenAI’s own models.
(UPDATED June 25, 2026 01:38 UTC)
OpenAI has announced the development of the Jalapeño chip, an Application-Specific Integrated Circuit designed to support future AI workloads, as they prepare for an initial public offering that could value the company at a staggering trillion dollars. In collaboration with Broadcom, OpenAI plans to deploy gigawatt-scale data centers alongside Microsoft and other partners starting in 2026, while their operational expenses for 2025 reached $34 billion, leading to an operating loss of nearly $20.92 billion. Additionally, OpenAI is set to utilize approximately two gigawatts of AWS's proprietary Trainium computing capacity over the next eight years, while Qualcomm has introduced the Dragonfly C1000, a new data center CPU intended for agentic AI, which Meta will implement when production begins in 2028.
--
(UPDATED June 24, 2026 21:23 UTC)
OpenAI has announced that early testing of its new Jalapeño chip indicates it will significantly outperform current state-of-the-art technology in terms of performance per watt. The engineering timeline for the Jalapeño chip has progressed rapidly from initial schematics to fabrication readiness in just nine months, and OpenAI is already testing its GPT‑5.3‑Codex‑Spark model on these chips. The collaboration with Broadcom aims to create gigawatt-scale data centers in partnership with Microsoft and others, with plans to begin deployment in 2026, all in an effort to make AI models more affordable and accessible to consumers.
--
(UPDATED June 24, 2026 17:40 UTC)
OpenAI and Broadcom have announced the development of the Jalapeño chip, which is designed to enhance the performance of ChatGPT and OpenAI's coding agent, Codex, and will be deployed across active data centers. OpenAI plans to begin rolling out these processors by the end of this year as part of a strategy to reduce reliance on existing chip companies like Nvidia and improve the affordability of its AI models. In 2025, OpenAI generated $13.07 billion in revenue but faced significant operational expenses, leading to an operating loss of nearly $20.92 billion, highlighting the need for more efficient chip solutions.