OpenAI’s Jalapeño Chip Shows Significant Performance Gains in Benchmarks
· Technology · TechCrunch, Business Standard, CNBC
OpenAI revealed benchmark results for its new Jalapeño chip at the Hot Chips conference, showing superior performance in tokens per user and throughput per kilowatt compared to current state-of-the-art inference processors. The chip, developed in collaboration with Broadcom, is designed to minimize delays during the prefill and communication phases of AI processing. OpenAI’s head of hardware, Richard Ho, stated that Jalapeño offers a significant performance advance over existing systems, including Nvidia’s Blackwell. Deployment is expected to begin in small volumes by late 2026, with broader availability in 2027. The chip is part of OpenAI’s multigenerational platform strategy, integrating AI products, models, and hardware.
Why it matters
This advancement could lower costs and improve efficiency for AI services, benefiting companies and end-users reliant on large-scale AI inference.
Read the original report — TechCrunch
Join us on Telegram
Breaking news the moment it lands. At 10,000 members we ship the Android app.