The term only exists because TypeSafe named its own category after Kahneman’s System 1. Its launch post cites “fast, intuitive System 1 thinking and slow, deliberate System 2 reasoning” as the source of the name. Once one half has a product label, people reach for the other half to describe everything that is not a System One model: chat models and reasoning models, anything that works in tokens and hands you prose.
Nobody sells a “System Two model”. If you see the phrase in a comparison table, read it as shorthand for a reasoning LLM.
The practical difference is time budget. TypeSafe claims Jev answers end to end in 70ms to 500ms, and that this is 40x to 200x faster than frontier models at the same level of intelligence on System One tasks. Both figures come from TypeSafe’s own evaluations; no independent benchmark had been published as of 2026-09-18. A reasoning model spends that budget generating intermediate text, which is what makes it better at multi-step problems and worse at sitting inside a request path that has to return in under a second.
The two occupy different slots in the same application. System One versus LLM works through where each one fits.