Perplexity is an AI-powered answer engine that performs real-time web searches and synthesizes findings into answers with inline citations, so every claim can be traced to its source. It offers a free tier with unlimited basic searches plus limited Pro searches per day, with paid plans unlocking unlimited Pro search, Deep Research, premium AI models, and file analysis.
Key features
Real-time web search with inline source citations
Pro Search for deeper multi-step research
Deep Research mode for comprehensive reports
Choice of frontier AI models (GPT, Claude, Gemini, Sonar)
File uploads and document analysis
Spaces for organizing research threads
Image generation capabilities
API access (Search API, Sonar API, Agent API)
Watch: Perplexity tutorial
Perplexity Full Tutorial — AI Foundations
Video by AI Foundations on YouTube — plays here without leaving this page.
Perplexity models, modes & features
No models or features match your search.
Models
Sonar (API, sonar)
ModelLegacy
Perplexity's base Sonar Chat Completions model. Perplexity's official migration guide maps Sonar to the Agent API 'fast' preset, described as best for single-fact lookups, definitions and quick summaries, and notes the Agent API is more accurate on research-style, multi-step questions.
Access
API developers (legacy Sonar Chat Completions)
Search-grounded answers with citations (API Platform is positioned as grounded LLM APIs)
Limits: Legacy: Perplexity's docs steer new projects to the Agent API; pricing no longer listed on the official pricing page
Higher-capability Sonar Chat Completions model. Perplexity's migration guide maps Sonar Pro to the Agent API 'fast' preset, described as best for everyday research questions and light multi-step lookups with current information.
Access
API developers (legacy Sonar Chat Completions)
Search-grounded answers with citations
Limits: Legacy: Perplexity's docs steer new projects to the Agent API; pricing no longer listed on the official pricing page
Reasoning-focused Sonar Chat Completions model. Perplexity's migration guide maps it to the Agent API 'low' preset, described as best for multi-hop browsing and wide aggregation across many sources.
Access
API developers (legacy Sonar Chat Completions)
Step-by-step reasoning over searched evidence
Limits: Legacy: Perplexity's docs steer new projects to the Agent API; pricing no longer listed on the official pricing page
Deep-research Sonar Chat Completions model. Perplexity's migration guide maps it to the Agent API 'high' preset, described as best for expert-level reasoning and exhaustive source coverage, often at lower per-request cost on the Agent API.
Access
API developers (legacy Sonar Chat Completions)
Exhaustive multi-source research reports
Limits: Legacy: Perplexity's docs steer new projects to the Agent API; pricing no longer listed on the official pricing page
Perplexity's own in-house model offered in the app's model picker for detailed answers with multiple references, per the official 'What is Perplexity Pro?' help article.
Access
Perplexity app (default in-house model; selectable on Pro)
Detailed answers with extensive sources and references
Limits: Picker availability can vary; help article last updated 2026-08-29
Reasoning ('Thinking') model listed in Perplexity's Pro help article for complex analytical questions that need to be broken down into multiple search steps.
Access
Pro subscribers (per official help article)
Breaks down complicated queries that require multiple search steps
Limits: Picker availability can vary; help article last updated 2026-08-29
OpenAI reasoning model that Perplexity's Pro help article lists as available to Max users rather than standard Pro subscribers, for complex analytical questions.
Access
Max subscribers only (per official help article)
Offered for complex analytical questions
Limits: Not available on standard Pro per the help article
Anthropic model that Perplexity's Pro help article lists as available to Max users rather than standard Pro subscribers, for complex analytical questions.
Access
Max subscribers only (per official help article)
Offered for complex analytical questions
Limits: Not available on standard Pro per the help article
Search modes: Best, Pro Search, Reasoning Search, Research
Mode
Perplexity's help article describes four search modes — Best, Pro Search, Reasoning Search, and Research — each paired with different available language models, and explains that users can switch between modes and models inside an existing session via the search box.
Access
All users; model/mode switching happens in the search box
Four modes matched to question typesModels can be switched mid-session
Perplexity's advanced search mode for nuanced, thorough answers to complex questions. Per its official help article, Pro Search synthesizes information from a diverse, high-quality set of sources, maintains context across follow-up questions, and cites original sources for every answer. Pro subscribers get extended access for multi-step reasoning, detailed research and complex projects.
Access
Pro subscribers get extended access (high volume of daily searches)
Multi-step reasoning searchesDetailed research and complex project queries
Limits: Free usage is limited (pricing page: Free is good for limited daily usage)
Perplexity's deep-research capability: Pro subscribers get extended access to Perplexity Research for deeper explorations, and the official pricing page lists 'Deep research' as a Pro feature and 'Expert level research' at the Max tier.
Access
Pro (extended access) and Max (expert-level research)
In-depth explorations beyond a single searchHighest tier described as 'Expert level research' on the pricing page
Limits: Free plan is positioned for simple questions and limited daily usage
Users can upload files — including PDFs, CSVs, audio, video and images — and explore their contents with AI models. The pricing page's feature matrix lists file uploads (Limited on Free, 10x on the paid tier shown).
Access
All plans (increased limits on Pro)
Analyze PDFs, CSVs, audio, video and imagesAsk questions about uploaded content with models such as GPT-5 and Claude Sonnet 4.6
Limits: Upload allowances differ by plan (Pro lists increased file and photo uploads)
Pro includes enhanced image generation and limited access to video generation tools using the latest image/video generation models, per the official Pro help article.
Access
Pro (enhanced image generation; limited video generation)
Enhanced image generation for visual research tasksLimited access to video generation tools on Pro
Limits: Video generation is limited on Pro; the pricing matrix shows higher video generation (5x) at the top tier
Pro includes access to Perplexity Create files and apps for experimental features and enhanced tools; the pricing page describes the paid tier as able to 'create polished documents & apps'.
Access
Pro and above (listed on the pricing page)
Create polished documents and apps from researchExperimental features and enhanced tools (Create)
Perplexity's agent/computer product line on the official pricing page: Pro includes expanded Computer access with 4,000 bonus credits; Max includes maximum Computer usage with 10,000 Computer credits per month.
Access
Pro (expanded access + 4,000 bonus credits); Max (maximum usage, 10,000 monthly credits)
Agentic task execution on top of Perplexity searchCredit-based usage with bonus credits on paid plans
Limits: Usage is credit-metered; plan allowances differ
Listed in the official pricing page's feature matrix alongside the Max plan: a mode that runs a query across multiple top models. The pricing page also lists 'Brain', 400+ app connectors and Comet Assistant among its advanced features.
Access
Shown on the pricing page with the Max tier
Cross-check one question across several frontier models
Limits: Positioned at the Max tier on the pricing page
consumer subscription (annual-billing price shown on the official pricing page; monthly-billing price not displayed on the page as fetched)
Deep research
Access to top AI models and the ability to select between AI models (e.g. Sonar, GPT-5.2, Claude Sonnet 4.6, Gemini 3.1 Pro; reasoning options Sonnet 4.6 Thinking and Grok 4)
Expanded Perplexity Computer access and 4,000 bonus credits
10x as many citations per answer
Extended access to Pro Search and Perplexity Research
Increased file and photo uploads (PDFs, CSVs, audio, video, images)
Enhanced image generation and limited video generation
Perplexity's API Platform (perplexity.ai/api-platform) offers an Agent API (the successor to Sonar Chat Completions, with built-in web search, URL fetching, sandboxes, MCP and premium finance/people search), a Search API for raw ranked web results, an Embeddings API and a Decisions API. The legacy Sonar Chat Completions models (sonar, sonar-pro, sonar-reasoning-pro, sonar-deep-research) map to Agent API presets (fast, fast, low, high) per Perplexity's official migration guide.
API pricing: Official API pricing page (updated 2026-10-20): Search API $5.00 per 1K requests; Search API with Fast Search $1.00 per 1K requests; Agent API models billed per token at each model's published rates, plus tool fees — web_search $0.0025 per invocation (Fast Search $0.001), image_search $0.0025, fetch_url $0.0005, people_search $0.005, finance_search $0.005, sandbox $0.03 per session (up to 20 active minutes); Embeddings from $0.004 per 1M tokens (pplx-embed-v1-0.6b); Decisions API $0.01-$0.02 per 1M input tokens with free output. Legacy Sonar per-token rates and per-request fees are NOT listed on the current official pricing page (see gaps). API usage is not included in consumer or Enterprise subscriptions — credits must be purchased separately.
Perplexity web app (perplexity.ai)Comet browser / Comet Assistant400+ app connectors (search across connected apps)Perplexity API Platform (Agent API, Search API, Embeddings API, Decisions API)MCP servers (via Agent API)
What should you use it for?
Quick factual questions with verifiable sources
Use standard search on Free for simple questions, or Pro Search for nuanced questions — every answer links to its original sources so facts can be verified.
In-depth research reports
Use Deep Research / Perplexity Research (extended access on Pro; expert-level on Max) instead of a single search when the topic needs many sources synthesized.
Questioning your own documents
Upload the PDF, CSV, audio, video or image and ask questions about it (increased upload allowances on Pro).
Choosing a frontier model per question
On Pro, pick the model per question — Sonar, GPT-5.2, Claude Sonnet 4.6 or Gemini 3.1 Pro for detailed answers; Sonnet 4.6 Thinking or Grok 4 for complex analytical questions.
Building search into your own product
Use the API Platform instead of a subscription: the Agent API for grounded, agentic answers, or the Search API ($5.00 per 1K requests) for raw ranked web results. API usage is billed separately and is not included in Pro or Enterprise plans.
Pros
Every answer includes checkable inline citations, making research verifiable rather than trust-based
Free tier is genuinely capable for everyday factual questions
Pro plan bundles multiple premium AI models into one $20/month subscription
Cons
Free tier caps advanced Pro searches per day (around 3/day), which regular users hit quickly
Synthesized answers still need verification against primary sources for high-stakes research
ai searchanswer enginecitationsreal-time webdeep researchfile analysismultiple ai modelsapi
Sources & verification
Model, plan and pricing information on this page was last verified on 22 October 2026 against official sources. Prices and line-ups change — always confirm on the official website before buying.
Legacy Sonar API per-token prices and per-request fees could NOT be verified from an official page: Perplexity's current official API pricing page (docs.perplexity.ai, updated 2026-10-20) no longer publishes a Sonar rate table — it covers the Agent API, Search API, Embeddings and Decisions API instead. Third-party sites quote Sonar at $1/$1, Sonar Pro at $3/$15, Sonar Reasoning Pro at $2/$8 and Sonar Deep Research at $2/$8 per 1M tokens plus request fees, but per the official-sources-only rule those figures are not recorded here.
Sonar context windows are null: no official page opened in this pass lists them (third parties commonly cite 128K for sonar / 200K for sonar-pro — unverified officially here).
Spaces and Labs are named in the task brief and the product UI, but no official Perplexity page opened in this pass describes them, so they are not listed as verified features.
Monthly (non-annual) sticker prices for Pro and Max were not displayed on the official pricing page as fetched — it shows annual-billing equivalents ($17/month Pro, $167/month Max). Third parties report $20/month and $200/month monthly billing from the same official pricing pages, but that was not captured from an opened page here.
Education Pro ($10/month for verified students/educators) is widely reported but was not verified on an official page opened in this pass, so it is excluded from plans.
The Pro help article's model list was last updated 2026-08-29; Perplexity rotates in-app models frequently, so the in-app picker may differ today. The article links to a 'full list of supported models' page that was not opened.
Perplexity Computer credit amounts for Pro beyond 'expanded access' plus the 4,000 bonus credits were not specified on the pages opened.
How to choose the right Education & Learning tool
A learning tool earns its place by how well it teaches, not how quickly it answers. Look for step-by-step explanations, worked examples and questions that check understanding, rather than finished homework on demand — the learning happens in the attempt. Match the tool to the level: school students need guardrails, simple language and safety controls, while university and professional learners need depth, sources and subject accuracy. Feedback quality is the dividing line: specific corrections on your work teach far more than generic encouragement. If the tool will be used by children, check age limits and privacy practices carefully, and prefer tools designed to guide practice instead of replacing it.
Perplexity is an AI-powered answer engine that performs real-time web searches and synthesizes findings into answers with inline citations, so every claim can be traced to its source. It offers a free tier with unlimited basic searches plus limited Pro searches per day, with paid plans unlocking unlimited Pro search, Deep Research, premium AI models, and file analysis.
What are the main features of Perplexity?
Perplexity offers Real-time web search with inline source citations, Pro Search for deeper multi-step research, Deep Research mode for comprehensive reports, Choice of frontier AI models (GPT, Claude, Gemini, Sonar), File uploads and document analysis, Spaces for organizing research threads, Image generation capabilities and API access (Search API, Sonar API, Agent API).
How much does Perplexity cost?
Perplexity follows a freemium model: a free tier is available, and paid plans unlock higher limits and advanced features. Check its official website for current plan prices.
What are the pros and cons of Perplexity?
Strengths of Perplexity include Every answer includes checkable inline citations, making research verifiable rather than trust-based; Free tier is genuinely capable for everyday factual questions; Pro plan bundles multiple premium AI models into one $20/month subscription. On the downside, Free tier caps advanced Pro searches per day (around 3/day), which regular users hit quickly; Synthesized answers still need verification against primary sources for high-stakes research.
Welcome to LLM Tools Hub! Hi! I'm Hubi 👋 I help you find the right AI tool from LLM Tools Hub. Tell me what you need — for example: “free AI tools for image generation”
