How long did it take to develop the chip?

The chip moved from design to production in just nine months.

How did Broadcom's stock react to the announcement?

Broadcom shares rose 3.4% on the morning of the announcement.

What is the significance of this chip for OpenAI?

The chip marks OpenAI's first foray into proprietary hardware, reducing reliance on external suppliers and giving it greater control over model performance.

Broadcom Rises 3% After Unveiling OpenAI 'Jalapeño' AI Chip

Broadcom (AVGO) unveiled 'Jalapeño,' a custom AI chip co-developed with OpenAI, sending shares up 3.4%. The chip, OpenAI's first proprietary 'Intelligence Processor,' was developed in just nine months.

June 24, 2026

2 min read

Source: Investing.com

Key Numbers

stock move

+3.4%

development time

9 months

Broadcom (NASDAQ:AVGO) shares rose 3.4% Wednesday morning after the company unveiled 'Jalapeño,' a custom AI accelerator developed in partnership with OpenAI. Billed as OpenAI's first 'Intelligence Processor,' the chip is designed specifically for large language model (LLM) inference and moved from design to production in just nine months.

The Product

'Jalapeño' marks OpenAI's official expansion into proprietary hardware as part of its full-stack infrastructure strategy. The chip is optimized for LLM inference, offering potential performance and efficiency gains over general-purpose GPUs.

Pricing and Availability

Pricing and commercial availability details have not been disclosed.

Competition

The chip enters a competitive landscape dominated by NVIDIA's AI accelerators, Google's TPUs, and Amazon's Inferentia and Trainium chips.

Potential Impact

For Broadcom, the partnership diversifies its revenue beyond traditional networking and telecom, tapping into the growing custom AI chip market. For OpenAI, it reduces reliance on external suppliers and provides greater control over model performance and cost.

Frequently Asked Questions

The 'Jalapeño' chip is a custom AI accelerator co-developed by Broadcom and OpenAI, serving as OpenAI's first proprietary 'Intelligence Processor' designed for LLM inference.

Found this useful? Share it

This article was rewritten in Wrqti's editorial style based on information from the original source above. Content is informational only — not investment advice.