OpenAI and Broadcom Unveil Jalapeño, an LLM-Optimized Inference Chip
OpenAI and Broadcom introduced Jalapeño, a custom AI chip built for LLM inference, per OpenAI's June 24 announcement.
At a glance
- OpenAI and Broadcom unveiled a custom LLM-inference chip named Jalapeño, per OpenAI's June 24 post.
- The stated goals are performance, efficiency and scale across OpenAI's AI systems.
VERDICT — CONFIRMED
OpenAI and Broadcom introduced Jalapeño, a custom AI chip built for large language model inference, per OpenAI's June 24 announcement.
The company says the chip targets improved performance, efficiency and scale across its AI systems, per the post. The announcement extends OpenAI's publicly stated push into custom silicon alongside partner Broadcom.
Technical and commercial specifics were not included in the material reviewed: no process node, no deployment volumes, no timeline for production or installation, and no cost figures accompanied the announcement.
The chip's existence and stated purpose are confirmed as company statements from OpenAI. Its performance claims are unverified — no benchmarks were cited in the material reviewed — and when Jalapeño will actually carry inference traffic at scale remains undisclosed by either company.
Key facts on file
- OpenAI and Broadcom unveiled a custom LLM-inference chip named Jalapeño, per OpenAI's June 24 post.
- The stated goals are performance, efficiency and scale across OpenAI's AI systems.
