OpenAI rolls out its 1st chip through a Broadcom tie-up as part of its “build the full stack” push.
Rohan Paul Twitter · Rohan Paul (@rohanpaul_ai) · 2026-06-24
OpenAI deploys its first custom AI chip, an ASIC called Jalapeño developed with Broadcom, offering lower cost and higher speed than Nvidia GPUs for fixed AI workloads as part of a full-stack infrastructure strategy.
Appears in
Extraction
Topics: custom-siliconai-chipsopenai-hardwarebroadcom
Claims
- OpenAI has rolled out its first custom chip, Jalapeño, developed in partnership with Broadcom.
- Jalapeño is an ASIC, making it less flexible than Nvidia GPUs but potentially cheaper and faster for well-defined AI workloads.
- The chip's architecture is designed to reduce data movement, a key bottleneck in AI inference.
- This chip is part of OpenAI's stated goal of building the full AI infrastructure stack.
Key quotes
Jalapeño is an ASIC, so it is less flexible than an Nvidia GPU, but can be cheaper and faster when the workload is known very well.
the architecture reduces data...