ChipsThe story, in brief
OpenAI, Broadcom debut custom Jalapeño chip for AI inference
OpenAI just stopped waiting for NVIDIA. Meet Jalapeño—the custom chip that changes the inference game.
The KeyNews take
Why it matters
OpenAI's move into custom silicon with Broadcom signals a critical shift: leading AI labs are now building their own hardware to reduce dependency on NVIDIA, cut costs, and optimize inference at scale. This is infrastructure consolidation, not a one-off experiment.
The key facts
5 to knowOpenAI + Broadcom partnership on custom Jalapeño chip for inference
Broadcom prior experience: designed Google TPU line
Chip targets LLM inference workloads
Signals vertical integration into custom silicon
Reduces reliance on third-party GPU suppliers
Go to the source
SiliconAnglesiliconangle.com
Publisher excerpt: OpenAI Group PBC today revealed a custom chip called Jalapeño that it will use to power its large language models. The processor is the fruit of a collaboration with Broadcom Inc., which is no stranger to custom silicon design. The company helped Google LLC develop its TPU line of artificial…