See Artificial Analysis to compare Ling 3.0 Tiny with other small open weights m
Source
artificialanalysis.ai
Author
ArtificialAnlys
Date
Context
Ling 3.0 Tiny's small active model size does not imply short generations. Artificial Analysis's comparison page lists 7.9 billion total parameters, with 1.3 billion active during inferenceRunning a trained model to get answers — the phase where AI is actually used, as opposed to trained.Full definition →, but also reports 230 million output 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 → across its Intelligence Index evaluation, above its comparison group's median.
That output count covers the evaluation suite, not one answer. It adds a separate dimension to model selection: the amount of text a model generates can matter alongside the amount of computation active at each step.
Terms in this piece · Glossary
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.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
token budget — A cap on how many tokens a task, session, or agent run may consume — the practical control on both cost and how long an agent will grind.
