Perplexity co-founder Aravind Srinivas launches Brain, a stateful memory system for the company's Computer product

With Brain, Computer starts each task with full context of your projects, decisions, and sources instead of from scratch.

On tasks that require past context, Brain improves answer correctness by 25%, recall by 16%, and runs 13% cheaper per task.

Every memory links back to the session, file, or source it came from with full transparency and control.

You can access Brain and your saved memories under "Customize" in the sidebar.

Read more about Brain: https://www.perplexity.ai/hub/blog/self-improving-memory-for-agents

@perplexity_ai A context graph that grows with every run needs to survive outside the session.

The sources behind each memory entry have to stay verifiable years from now.

That's a storage problem, not just a memory one.

@perplexity_ai Memory inside Perplexity is a real step. Walrus Memory plugs in when the same agent needs memory across Perplexity, Claude, Codex, and Cursor. 🦭

@perplexity_ai 🔥🔥

@perplexity_ai Connectome

@perplexity_ai Incredible.

@perplexity_ai Brilliant, thank you.

@perplexity_ai perplexity after every few days

@AravSrinivas Does Brain allow me to reference work from one thread in another without having had to prompt “add this to memory”?

@perplexity_ai @bughuntergeek I already built this in FreeLattice.

Anyone who wants the code, it’s open at GitHub-> chaos2cured -> FreeLattice

Also, mine also has autonomous building.

Perplexity is less evil than OpenAI. I hope you all continue to build and fight for everyone to have equal access. •

@perplexity_ai Perplexity is useless

@perplexity_ai Impressive vision—persistent memory could make AI workflows far more contextual.

@perplexity_ai That's awesome

@perplexity_ai Max max max, es una pena..no aporta algo relevante para los usuarios que tienen resuelto un arnés propio con ese tipo de construcción o incluso potenciada, posiblemente sea "novedoso" para quienes no trabajen su propio sistema de memoria..

@perplexity_ai @grok is this agi ?

@AravSrinivas I think computer is great, but it is expensive… can you add perplexity model agent or allow OpenRouter ???

@perplexity_ai The lift's real but the baseline is the story: correctness was 0.497 without Brain, basically a coin flip, and Brain only gets it to 0.622.

@perplexity_ai

@perplexity_ai How do you use it or is it just “on” by default?