A guide · 6 min read
Perplexity isn't built like ChatGPT, and its memory model reflects that. It's an answer engine first — research-shaped — and its persistence features (Threads, Spaces, account-level history) follow that grain. Here's what each one actually does and when to reach for which.
Short answer
Perplexity remembers within a Thread (a continuing research conversation) and within a Space (a curated project folder with pinned docs and instructions). It does not auto-learn ambient facts about you across threads. For ambient memory or cross-provider continuity, you need a separate layer.
A Thread is a continuing research conversation. Follow-up questions stay in context, and you can return to the same thread days later to keep digging. Threads are visible in your Library and persist as long as your account does. They do not, however, share context with other threads — starting a new Thread is a fresh slate.
Spaces (Perplexity's evolution of Collections) are project-scoped containers. You can pin source documents, set custom instructions, and group related Threads under one Space. This is the closest Perplexity gets to Claude Projects: explicit, curated context that persists across the queries you run inside it.
In Settings → Profile you can store information about yourself (role, interests, preferences) that gets factored into responses. This is the closest thing Perplexity has to ChatGPT Memory, but it is a single block of text you maintain manually — not an auto-learning system.
Konshus is one way to solve this — a persistent memory vault and portable persona that follows you across ChatGPT, Claude, Gemini, and whatever ships next.
Meet KonshusKonshus is one way to solve this — a persistent memory vault and portable persona that follows you across ChatGPT, Claude, Gemini, and whatever ships next.
Meet Konshus