Jalapeño’s first results show industry-leading speed and efficiency in AI inference

OpenAI now has its own inference accelerator with published results, which bears on where serving capacity, latency and cost for its models are heading.

• inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.

• postOpenAI's first inference chip posts results, deploys by year-endOpenAI
• articleOpenAI Jalapeño: Better Than Nvidia Blackwell OpenAI’s self-designed ASIC…SemiAnalysis
• postOpenAI's first in-house AI chip, Jalapeño, reaches productionSam Altman

Checking sign-in…

Loading comments…