OpenAI’s Jalapeño chip is built for fast inference at scale, benchmarks show

August 25, 2026 Russell Brandom

Tested on SemiAnalysis’ InferenceX benchmark, Jalapeño registered both more tokens per user and more throughput per kilowatt than the currently available state-of-the art.

Previous Article
Apple rescues Hide My Email feature from the privacy scrap heap
Apple rescues Hide My Email feature from the privacy scrap heap

Apple says it will no longer ditch using its icloud.com domain for hiding people's email addresses.

Next Article
Apple debuts its ‘most powerful chip ever’ in M5 Ultra and M6
Apple debuts its ‘most powerful chip ever’ in M5 Ultra and M6

Apple unveils these new processors alongside an updated Mac Mini and Mac Studio.

LINK CTA

Learn More