OpenAI tests first custom inference chip Jalapeño, boosts efficiency and speed

OpenAI · x · 2026-08-26

OpenAI announced results for its first custom inference chip, Jalapeño. Tests show a major advance in intelligence per watt and response times, delivering higher throughput and lower latency without sacrificing efficiency. This means faster ChatGPT responses, more responsive Codex sessions and agents, and reliable access amidst growing demand.

Related event: OpenAI Claims Custom Inference Chip Jalapeño Beats Nvidia Blackwell(34 posts)→

Original post →

More from Infra

Infra channel →