ainewsblitz.com

Breaking

OpenAI Reports Jalapeño ASIC Delivers Up to 1.9 Times More Inference Work per Watt Than Nvidia Systems

  • Infra & Chips
  • AI Agents

OpenAI has published the first measured results for Jalapeño, the inference ASIC it designed with Broadcom, saying the chip completed 1.5 to 1.9 times more AI work per watt at peak throughput and returned answers 1.7 to 3.6 times faster end to end than Nvidia GB200 and GB300 systems. On highly interactive workloads, the company reported 2.1 to 4.1 times higher performance.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,528 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year