The three models
Core 2, Core 3.1 and Logos side by side — what each is for and how to switch.
Socria ships three models behind one switcher, and they are not three sizes of the same thing — they are three different amounts of machinery around the conversation. Core 2 is a voice. Core 3.1 is a voice with a memory and a running read of the thread. Logos is an environment.
Side by side#
| Core 2 | Core 3.1 | Logos | |
|---|---|---|---|
| In a sentence | Calm, restrained Socratic questioning in plain prose | Assertive pattern-naming with thread memory and adjustable depth | The conversation plus a live map of your reasoning |
| Account | Not required — one free session signed out | Sign-in (or an access key) | Sign-in (or an access key) |
| Thinking Depth control | — | All four modes | Its own depth control; all four with One |
| Memory | None beyond the visible thread | Thread memory, syntheses, insights, a cross-conversation journey | The map itself — plus saved lines of thinking |
| Extra surfaces | — | Synthesis & insight cards, choice chips | Thinking Map, Board, plots, Draft Space, Research |
| Writes prose for you | Only refining material you brought | Only refining material you brought | Never — even Draft Space’s Refine is a proposal that lands only when you apply it |
Switching between them#
The model picker sits bottom-right, beside the chat box. Picking Logos does not navigate anywhere: the whole surface swaps in place inside /chat, because Logos is a model, not a destination — and leaving Logos returns you to whichever Core model you were on before, not to a default.
Your choice is remembered per browser. In the sidebar, chat and Logos sessions sit in one list ordered by when you last touched each; Logos sessions carry the Logos mark, and opening one switches you into Logos. The way back is the Socria chat button in the Logos header, which returns you to whichever Core model you were on.
Which one, when#
- Core 2 — you want questions, not machinery. It is also the model that works with no account at all.
- Core 3.1 — a thread you will return to. It notices your language, names patterns without hedging, asks at most one question per turn, and periodically hands you a structured synthesis of what you have actually worked out.
- Logos — thinking with structure: decisions, learning, math, anything where seeing the reasoning matters as much as having it.