We’ve designed and built our first AI chip: Jalapeño. —San Altman CEO @OpenAI Designed from the ground up by
OpenAI announced Jalapeño, its first custom AI inference chip, designed by OpenAI and produced with Broadcom to accelerate LLM workloads that power ChatGPT, Codex, the API, and future agentic products. The chip is part of a multi‑generation compute platform intended to improve performance, efficiency, and scalability across OpenAI’s full‑stack offering, helping the company serve more users and expand access to AI [openai]. Early coverage notes the chip is a reticle‑sized ASIC developed rapidly and claimed to deliver high performance per watt for modern LLM inference [tomshardware].
Follow-up Questions:
1. How does Jalapeño’s architecture differ from GPUs and TPUs?
2. When and how will Jalapeño be deployed across OpenAI’s services?
3. Will third parties be able to buy or license Jalapeño-based hardware?
4. What performance or efficiency gains has OpenAI published for Jalapeño?
5. How does this affect cloud providers and existing AI hardware vendors?
Sources
- OpenAI and Broadcom unveil LLM-optimized inference chip | OpenAI
- OpenAI on X: "We’ve designed and built our first AI chip: Jalapeño. Designed from the ground up by OpenAI and brought to production with @Broadcom, Jalapeño is purpose-built for the LLM workloads powering ChatGPT, Codex, the API, and future agentic products. Chips are foundational to the AI https://t.co/mHU7DaMMTi" / X
- Broadcom and OpenAI debut Jalapeño Intelligence Processor, plot an Apple-like move to 'build the full stack' | TechRadar
- OpenAI's First Custom Chip Is As Hot As A Jalapeño For AI, As The Firm Calls It The "Best Inference Platform" for LLMs
- Broadcom and OpenAI unveil custom-built Jalapeño inference processor — OpenAI's first chip is a massive reticle-sized ASIC built in an ultra-fast nine-month development cycle | Tom's Hardware
Related questions
- How does Jalapeño’s architecture differ from GPUs and TPUs?
- When and how will Jalapeño be deployed across OpenAI’s services?
- Will third parties be able to buy or license Jalapeño-based hardware?
- What performance or efficiency gains has OpenAI published for Jalapeño?
- How does this affect cloud providers and existing AI hardware vendors?