Perplexity
Updated May 2026
Perplexity
The AI-native search engine
Perplexity isn't primarily a model maker — they're an AI search product that wraps other labs' models (Claude, GPT, Gemini, Grok) behind a single interface with strong citations. Their own Sonar models are fine-tuned for grounded web search. In 2025 they launched Comet, an AI-native browser, which became free in 2026 and now bundles their Model Council feature — run the same query across multiple frontier models side by side.
Variants
5
Best at
unmatched citation UX
Max context
200K
tokens
Pricing
Free
$0
per limited queries
Pro
$20
per month
Enterprise Max
$325
per user/month
Sonar API
$1 in / $1 out
per million tokens + $5/1000 searches
Variants
Sort the table by context window or cost to find the right variant. Click any version below for a battle card with ranks, pricing notes, and official links.
| Modalities | ||||
|---|---|---|---|---|
Sonar Pro sonar-pro | 200K | varies / varies | 2025 | text |
Sonar Deep Research sonar-deep-research | 200K | varies / varies | 2025 | text |
Sonar sonar | 128K | varies / varies | 2025 | text |
Agentic Research API pplx-agentic | — | varies / varies | 2026 | text |
Comet Max comet-max | — | varies / varies | 2026 | textvision |
Battle card
Context rank
#1
within Perplexity
Capability rank
#3
modalities + reasoning
Weights
Closed
API/app access
Best fights to pick
- citation-required research
- news summarization
- fact-checkable Q&A
Rankings are Tendril directory ranks, computed from the model data shown here. Public benchmark leaderboards change often, so official docs and current benchmark pages should be checked before buying or deploying.
Learn
Lessons about this model
Structured lessons that cover Perplexity directly or put it in context alongside its rivals.
Perplexity Sonar — when search-first beats raw reasoning
Perplexity: The AI Answer Engine That Replaced Google For Many
Perplexity Comet — the AI browser
Perplexity Spaces — your personal research lab
Perplexity for Real-Time Research
Gemini Deep Research — autonomous research pipeline
Check yourself
Quizzes
Short, mixed-difficulty quiz sets on Perplexity and its model family.
Specialized Models
4 questions
Beyond general chat — coding, math, embeddings, and domain models.
Start quiz →Hands-on
Try these prompts
Ready-made prompts that show Perplexity at its best. Use them in your own AI workspace, then compare the output with what you learned in Tendril.
Perplexity vs. Google on a fresh question
ExplorersAsk something time-sensitive where citations matter — Perplexity's strength is sourced answers.
What were the key announcements at yesterday's Apple event? Give me a bulleted summary with citations for each claim, and flag anything that's still rumor vs. officially confirmed.
Sonar for programmatic search
CreatorsUse Sonar's API to get cited answers you can embed in a product.
For each company in this list, find their current CEO (2026) and the source page confirming it. Return as JSON: { company, ceo, source_url }. [List]Spaces for a longer research thread
CreatorsSet up a Space with a focused research question and iterate over days.
I'm researching the history of zoning reform in US cities 2015-2025. Help me build a reading list of the 15 most-cited papers and articles, then summarize each in 3 bullets. Add new sources as I paste them.
Print & keep
Printable reference
One-page summaries and flowcharts — great for desks, classrooms, or study sessions.
Go deeper
Official resources
Straight from the lab — docs, API references, and the chat surfaces you can try today.
Strengths
- unmatched citation UX
- Comet AI browser
- Model Council for comparing LLM answers
Limits
- own models trail frontier labs
- business model depends on wrapping others' APIs
- search adds cost per query