OpenAI's custom inference chip posts record efficiency in early benchmarks
OpenAI presented benchmark results for its Jalapeño inference chip at Hot Chips, showing it delivers higher tokens per user and throughput per kilowatt than current state-of-the-art processors like Nvidia's Blackwell. The chip, developed with Broadcom, is designed to minimize delays in prefill and communication phases. Deployment is expected in small volumes by end of 2026, with larger scale in 2027.
Related stories
OpenAI claims its new AI chip beats Nvidia on inference benchmarks · Semiconductors
Intel's Crescent Island AI chip focuses on inference efficiency with Xe3P architecture · Artificial intelligence
This summary is AI-generated and original to Mobble; the linked article is the authoritative source.
Original headline: “OpenAI’s Jalapeño chip is built for fast inference at scale, benchmarks show.” Browse more stories.