OpenAI Jalapeño: Better Than Nvidia Blackwell OpenAI’s self-designed ASIC…
Source
newsletter.semianalysis.com
Author
SemiAnalysis
Date
Why it matters
A first-generation custom inferenceRunning a trained model to get answers — the phase where AI is actually used, as opposed to trained.Full definition → ASIC beating Blackwell on tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition → per megawatt without speculative decodingA speed trick where a small model drafts several tokens ahead and the big model verifies them in one pass, often doubling generation speed.Full definition → or prefill-decode disaggregation changes the assumption that inference cost curves are set by NVIDIA's roadmap.
Terms in this piece · Glossary
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
speculative decoding — A speed trick where a small model drafts several tokens ahead and the big model verifies them in one pass, often doubling generation speed.
