OpenAI still isn’t giving up Nvidia chips, though.
128 chips, 1.7 exaFLOPS, and 27 TB of HBM give Altman and crew a leg up over Blackwell, and maybe even Rubin
The chip is designed for serving up AI models, rather than training new ones.
OpenAI showed off "Jalapeño," its first in-house inference chip, with benchmarks at the Hot Chips conference. According to SemiAnalysis tests, the chip beats Nvidia's Blackwell and even Rubin in throughput and energy…
Tested on Semianalysis’s InferenceX benchmark, Jalapeño registered both more tokens per user and more throughput per kilowatt than the currently available state-of-the art.
La tecnológica OpenAI aseguró que su nuevo chip de inteligencia artificial (IA), bautizado como Jalapeño, supera a los procesadores de Nvidia.
OpenAI reports its Jalapeno chip outperformed Nvidia's GB300 on power efficiency and response speed, in tests it conducted and against a superseded part.
Co-designed with OpenAI's own AI models in a rapid nine-month cycle from architecture to tapeout, Jalapeño demonstrated higher throughput per watt and lower latency when compared…
OpenAI's custom Jalapeño inference chip outperformed Nvidia's GB300 processors by up to 3.6x in latency tests, with a planned rollout beginning
OpenAI has introduced Jalapeño, its OpenAI’s first Intelligence Processor, as a purpose-built...
OpenAI showed off "Jalapeño," its first in-house inference chip, with benchmarks at the Hot Chips conference. According to SemiAnalysis tests, the chip beats Nvidia's Blackwell…
Nvidia's Vera Rubin wasn't in the comparison.
Exactly who at OpenAI decided to name the company’s first internally designed chip Jalapeño? That might have seemed like a good idea a few months ago—perhaps OpenAI imagined that…