Perplexity Spaces
Perplexity's shared folders for recurring research projects
Overview
Spaces group Perplexity threads, uploaded files and standing instructions into one project workspace, so context persists across sessions instead of resetting each time. Over 5 million Spaces existed by March 2026.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 11 Jul 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- 50 files per Space (25MB each)
- Deep Research + Labs
- Faster, higher-usage models
- 500 files per Space
- SSO-integrated permissions
- Queries excluded from training
- All Enterprise Pro features
- Expanded performance/model access
Use cases
What you can produce with Perplexity Spaces
- Grouped threads, files and custom instructions per project
- Up to 50 file uploads per Space on Pro (500 on Enterprise)
- Computer agent can create/edit docs and sheets inside a Space
- Named collaborator invites with granular permissions
- SSO integration and training-data exclusion on Enterprise Pro
- Shareable, persistent context across recurring research threads
ASEAN Perspective
Perplexity Spaces in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
Spaces solve a real problem with chat-based research tools: without them, every Perplexity thread is an island, and repeating context for a running project gets old fast. A Space bundles the threads, uploaded files (up to 50 files/25MB each on Pro) and a standing instruction set for one project, so a quarterly competitive review or a client account can accumulate context instead of starting over each time. Computer, Perplexity's newer agent, can now create and edit docs inside a Space and remembers its progress across sessions.
The catch is accuracy, not organization — Perplexity's failure mode is confident wrongness, not silence, and users report it missing on math problems a middle schooler would get right. Spaces don't fix that; they just make the wrong answer easier to find later. Worth it for teams running multiple concurrent research threads on Perplexity already; not a reason to pick Perplexity over a competitor on its own.
What people say
Perplexity carries a 4.4-star rating across 326 reviews on G2 (an older, smaller sample put it at 4.7 from 47 reviews — the larger, more recent number is the one worth trusting). Spaces specifically get credited by reviewers for letting them build "custom agents with specific instructions for any topic" and combine Search, Deep Research and other tools inside one project container instead of starting fresh each session — several describe it as saving real time by doing the initial legwork of gathering sources before they even open a browser tab.
The recurring complaint isn't about Spaces as a container — it's about what goes inside it. Perplexity doesn't reliably flag its own uncertainty: "when Perplexity is wrong, it doesn't know that it's wrong," as one review analysis put it, and users report basic math errors slipping through unflagged. That matters more inside a Space than in a one-off query, because a Space is meant to accumulate trusted context over weeks, and a bad answer baked into the file set doesn't announce itself.
Mobile is a separate sore point: the Android app has had persistent complaints about PDF handling, with some users reporting it won't read past roughly the first 10 pages of a longer document — a real problem for a workspace feature explicitly built around file uploads. There are also reports of the assistant losing thread of a recent exchange, which cuts against the pitch of a Space as persistent memory.
On the collaboration side, Perplexity added real permissioning in 2026 — space owners can invite named collaborators, and the Enterprise Pro tier ties that into SSO so internal research stays inside the org, with files and queries excluded from model training by default. That's a meaningful upgrade for regulated teams (legal, finance) who were previously locked out of using Spaces for anything containing real client material. Free-tier access to Spaces is limited; the file-heavy, multi-collaborator version of the feature is really a Pro-and-up proposition.
Summary of public user & expert reviews, compiled by RECATOOLS.
About this listing
This entry was compiled from publicly available data including Perplexity Spaces's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Perplexity Spaces unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Perplexity Spaces directly →
Spotted something out of date? Suggest an update →
More in Research & Data
