Perplexity

1. Should you try Perplexity? Overview

What it is	Perplexity is an answer engine that fuses real-time web search with frontier LLMs, returning cited, verifiable answers plus Deep Research reports, an agentic cloud computer and the Comet browser.
Country / HQ	San Francisco, USA
Best for	Journalists / investors / policy & market researchers needing auditable real-time intelligence
Check before buying	Watch out: Long-horizon academic PDF/literature synthesis (Gemini 2.5 Pro 2M ctx deeper)
Pricing signal	Free tier + paid from $5/mo

2. Features What it does

Feature	What it does
Answer engine	Real-time web index, Sonar (own) + frontier models, citations on every claim, visible reasoning chain, follow-up threading — Open-web only, no gated DB
Deep Research	40+ sources, ~4 min, 90%+ citation accuracy, structured report — Free 5/day; Pro caps changed twice in 2026
Multi-model	~20 models orchestrated: Claude Opus 5 (reasoning/Computer lead), GPT-5.6 (long ctx/recall), Gemini (deep research), Sonar 2, Grok (speed), Nano Banana (img), Veo 3.1 (video) — No single model >25% Dec 2025
Labs	Generate polished docs, apps, data assets, code — Pro+
Pages / Spaces	Pages = formatted shareable report; Spaces = team research desk + custom instructions — Pro+
Brain	Self-improving memory wiki (people/projects/prefs), 15 langs, ~90% Max users, +10pp memory benchmark — Max only, global Jul 2026
Computer	Cloud AI agent, Firecracker VM isolation, decomposes objective → sub-agents, codes/generates files/remembers, Slack @computer — Consumer Feb 2026 → Enterprise Ask 2026; 100+ connectors, custom API creds vault
Comet	Chromium browser, answer-bar + sidecar, live-tab agent actions, cross-tab summary, voice, Finance Hub, ad/tracker block, Model Council (3 LLM parallel) — Free since Mar 18 2026; no Linux
Enterprise	SSO/SAML, SCIM group sync, RBAC + custom roles + group credit limits (Jul 26 2026), audit, SOC2 Type II, zero-retention option, business connectors, workflow templates (legal/finance/sales/support), org-wide credit pool — Annual sales contract
Publisher	Comet Plus $5: FT/WSJ/Condé Nast/CNN/Fortune/WaPo/Le Monde partner content — Included Pro/Max

3. Pricing Pricing (2026-10, USD)

Plan	Price / mo	Key limits & who it fits
Free	$0	$0 Unlimited quick searches (Sonar), 5 Pro searches/day, basic models, search history, no file upload, no image gen, no Labs/Pages/Spaces/Brain, no Computer
Pro	$20	$200/yr ($16.67) 300+ Pro searches/day, GPT-5.6/Claude Sonnet 5/Sonar 2, Deep Research (caps revised 2026), file uploads, image gen, Labs, Pages, Spaces, $5 API credits/mo, Comet included
Max	$200	$2,000/yr Unlimited advanced models, 10,000 bonus credits/mo, Claude Opus 5, Deep Research headroom, Labs/asset gen 5×, video 5×, Computer, Brain, Model Council, priority, early access
Education	$10/mo	— Pro features, verified student
Enterprise Pro / Max	Custom	Annual contract RBAC/custom roles, SCIM group sync, group credit caps, SSO, audit, zero-retention, business connectors, workflow templates, org credit pool
Comet Plus	$5/mo	— Premium publisher content; free with Pro/Max
API	PAYG	— Sonar: $5 per 1M input tok / $2 per 1M cached; Enterprise tier available

4. Integration Access & integrations

Access type	Details
Access	access_type: web-app-plus-mobile-plus-browser-plus-api-plus-slack access_display: 🌐 perplexity.ai (search/spaces/labs/pages/computer) · 📱 iOS/Android · 🖥️ Comet browser (Win/Mac/iOS/Android, Chromium, no Linux first-party) · 🔌 API (Sonar) · 💬 Slack @

5. Reviews & Sources External reviews

Source	Rating	What it says
AI Stack Digest (7-tool test)	Deep Research: 40-source market analysis in ~4 min vs 2 hrs human, citation accuracy 90%+; real-time freshness decisive for journalists/investors; weaker on scientific PDFs vs Gemini
FuturePicker	“Same $20, 30× quota gap: Perplexity Pro 20 Deep Research/month, Gemini AI Pro 20/day (600/mo); free tier flips it — Perplexity 5/day vs Gemini 5/month”
VentureBeat (Ask 2026)	Computer = “single biggest productivity unlock in company history”; 100+ enterprise customers in one weekend; orchestrates 20 models, Firecracker-isolated
AI Agent Square / RECATOOLS (Comet)	8.4/10 — “most polished consumer AI browser 2026, free removed only objection”; power users need Max $200 for Background Assist/Labs/Computer
ToolChase / AI Test Guide	Pro worth it if research is daily work; warning: recurring friction on subscription downgrades + support response times, test monthly before annual
Security research (early 2026)	Comet blocked only small fraction of phishing vs Chrome; MCP-exploit class lets malicious servers push code through agent — don’t grant sensitive-tab access

6. Best for / Not for Fit

✔ Suitable

✘ Not suitable

• Journalists / investors / policy & market researchers needing auditable real-time intelligence • Analysts doing competitive + pricing + regulatory mapping at speed • Students (Education $10) learning to verify sources • Teams wanting Spaces as a shared research desk • Power users running Computer workflows (Max $200) • Comet users (free Chromium) who want assistant-in-tab • Enterprise with Slack-first ops + need SSO/SCIM/zero-retention

• Long-horizon academic PDF/literature synthesis (Gemini 2.5 Pro 2M ctx deeper) • Creative writers / open-ended chat (ChatGPT/Claude stronger) • Orgs relying on Bloomberg/IEEE gated data (keep terminals) • High-stakes finance/legal acting on citations without manual check • Privacy-strict orgs uncomfortable with agent reading every tab (Comet data handling) • Linux-first users (no first-party build), non-technical users on sensitive browsing • Companies averse to copyright/opt-out controversy

7. Competitors How it compares

Competitor	How it differs	Best for
Google Gemini Deep Research	2M ctx, narrative depth, 30× Pro quota, Workspace-grounded	$0 / $19.99 AI Pro
ChatGPT Deep Research + Operator	Broad generalist + agent, weaker citation heritage	$20/mo Pro
Microsoft Copilot	Graph/Office-grounded, $30/seat add-on	$0 / $30+seat
You.com / Andi	Search-native chat, privacy angle	Free / paid
Elicit / Consensus	Academic/source-grounded research	Free / $49-mo
Semantic Scholar	Free paper discovery + API	$0
Neeva / Brave Search	Independent search, weaker AI agent	Free
Iris.ai / Causaly	Regulated R&D Agentic RAG	Enterprise custom

8. Summary Verdict

Who should pick it	Journalists / investors / policy & market researchers needing auditable real-time intelligence
Who should skip it	Long-horizon academic PDF/literature synthesis (Gemini 2.5 Pro 2M ctx deeper)