mixture-of-experts — A model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.
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.
open weights — A model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.
inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
Why it matters
Trinity Large ships as open weightsA model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.Full definition → at 400B total and 13B active, with 4-of-256 routing that is sparser than most peers and roughly 2-3x faster inferenceRunning a trained model to get answers — the phase where AI is actually used, as opposed to trained.Full definition → on the same hardware. Base and TrueBase checkpoints are released alongside the chat-ready Preview.
