Overview
Perplexity is an AI answer engine: every answer comes with citations to its sources. It routes questions to models from several providers (OpenAI, Anthropic, Google and others) plus its own Sonar models, and paid plans let you choose a model.
Agents use it for research: local regulations, market questions, vendor comparisons and quick fact-finding with sources they can check.
Key features
- Cited answersSources on every answer.
- Model choicePick from leading models on Pro and Max.
- Deep ResearchLonger research reports.
- File uploadsAnalyse documents.
- VoiceAsk questions by voice in the apps.
Who it's for
Best for
Agents who research a lot and want sources.
Skip it if
You mainly need long-form writing.
Pricing
| Plan | Price | What you get |
|---|---|---|
| Free | $0 | Basic models, limited Pro Searches. |
| Pro | $17/mo billed annually | Top models, model selection, Deep Research. |
| Max | $167/mo billed annually ($200 monthly) | Frontier models, highest limits. |
Free and paid model lineups change often. Prices checked on the vendor's pricing page in Oct 2026; confirm before you buy.
Scores, pros & cons
Score breakdown
Pros
- Citations on every answer
- Choose between leading models on Pro
Cons
- Weaker for long-form writing
- Answers still need checking
Integrations
Connects to files and apps on paid plans.
Ease of use
Very easy.
Bottom line
The best assistant for cited research. Pair it with Claude or ChatGPT for writing.
Visit Perplexity →FAQ
Does Perplexity cite sources?
Yes, every answer includes citations.
Which models does it use?
Models from several providers plus its own Sonar.
Is it free?
Yes, with limits.
How much is Pro?
$17/month billed annually.