Meet Jalapeño: OpenAI’s First Custom AI Chip, Built With Broadcom
Broadcom says Jalapeño can run inference at roughly half the cost of a typical AI GPU, though the figure has not been independently benchmarked.
- On June 25, 2026, OpenAI unveiled a proprietary ASIC chip developed with Broadcom to optimize inference workloads. This development marks a shift in how the company manages artificial intelligence processing.
- OpenAI designed the processor to reduce reliance on general-purpose hardware. This strategic shift allows the company to control infrastructure from the model level down to the silicon layer.
- As an ASIC, the chip targets memory bottlenecks that typically slow large language models. The design pairs a compute section with stacks of high-bandwidth memory to optimize data movement and reduce power consumption.
- Regarding cost, Broadcom executive Hock Tan claims the processor performs inference at reduced costs compared to a standard GPU. This estimate stems from internal workloads chosen by Broadcom rather than independent benchmarks.
- Taiwan Semiconductor remains critical for OpenAI, as the company must compete for finite fabrication allocation alongside other technology firms to build the new chips.
12 Articles
12 Articles
OpenAI pulls ahead on custom chips
Since the ChatGPT moment nearly four years ago, the AI race has been viewed by many as a contest of model performance. Say what you will about OpenAI co-founder and CEO Sam Altman, but he saw from the very beginning that compute infrastructure would become an important battlefront, and his company is getting ahead in a key element that can make models do more, faster and more cheaply. His long-term strategy was evident earlier this week when Ope…
Why everyone from OpenAI to SpaceX is building their own chips (and turning up the heat on Nvidia)
Nvidia has dominated the AI chip market for years, but the era of total dependence might be ending. OpenAI just shared its plans to spice things up with Jalapeño, its custom inference chip built with Broadcom, joining Google, Apple, and SpaceX in a growing list of companies building their way out of single-supplier risk. The goal is less of a [ . . . ]
OpenAI presents Jalapeño, its first home AI chip designed with Broadcom. The idea is simple, make ChatGPT run faster and at lower cost.
OpenAI Unveils Jalapeno AI Inference Chip Built with Broadcom - Tech
OpenAI has unveiled Jalapeño, its first custom-built AI inference chip, developed in partnership with Broadcom, reducing dependence on third-party chipmakers. With AI systems becoming more advanced and attracting more users, demand for specialised hardware built specifically for AI workloads is increasing. OpenAI is not alone in this transition as Meta also advanced its own in-house silicon ambitions through its Meta Training and Inference Accel…
Coverage Details
Bias Distribution
- 34% of the sources lean Left, 33% of the sources are Center, 33% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium
