TechOpenAI’s Jalapeño chip built for fast inference at scale, benchmarks show
On SemiAnalysis’ InferenceX benchmark, OpenAI’s Jalapeño recorded both more tokens per user and higher throughput per kilowatt than the best state-of-the-art systems currently available.
T
TechCrunch★★☆☆☆2026-08-25 14:22Original
OpenAI’s Jalapeño chip built for fast inference at scale, benchmarks show
Tech