Text model

Perplexity Sonar — answers with links to the sources

Q: Which Perplexity models are available in GPTunneL?

The current list is always visible in the «Every Perplexity model» block above — it is built straight from the platform catalogue.

Q: How is Sonar different from a regular model with web search?

For Sonar, search is not an optional extra but the way it works: the model builds the answer around live results from the start and always attaches its sources. A regular model goes online only when it decides to.

Q: What is Deep Research?

A model that works a topic over many steps — it writes its own sub-queries, visits dozens of pages and assembles a full review. The answer takes longer to prepare, but the coverage is far wider.

Q: Can I trust the links in the answer?

The links point to the real pages the answer was built from, and they are worth opening. The model can paraphrase a source imprecisely, so for anything important check the original.

Q: Which languages does Perplexity handle?

The models understand the request and answer in the same language, and can search sources in that language too.

Sonar models by Perplexity in GPTunneL: the answer is built from a fresh web search and arrives with its sources. Pay per token, no subscription, no VPN.

Try Perplexity Available models

What the Sonar family is good at

The traits Perplexity keeps from generation to generation — you can count on them whichever model you pick.

Answers from a live search

The key difference from an ordinary model: Sonar does not rely on memory. It goes to the web and builds the answer from what is there right now, so the training cutoff stops mattering.

Links to the sources

The answer comes with the list of pages it was assembled from. A fact can be checked straight away, with no guessing where a number came from.

Deep research

A separate Deep Research model works a topic over many steps — it writes its own sub-queries, visits dozens of sources and assembles a full review.

Reads images

The larger models understand an attached screenshot or photo and factor it in alongside the search.

Which model to pick

Three models in the family — for different depths of research.

	SonarQuick questions	Sonar ProThe default pick	Deep ResearchA full review
Best for	Checking a fact, a rate or a news item fast	Working a topic with sources, comparing options	Multi-step research across dozens of sources
Sources visited	A handful	More, and more varied	Dozens
Speed	Fastest	Fast	Slow — the answer takes a while
Price per token	Lowest	Above average	Medium

Specs come from the platform catalogue. Exact per-model prices are on the Pricing

Every Perplexity model

Release dates follow Perplexity's official announcements.

March 2025

Sonar Deep Research

The model for multi-step work: it breaks a topic into sub-queries on its own, visits dozens of sources and assembles them into a full review with links.

February 2025

Sonar Pro Sonar

The first Perplexity models available outside its own search product: the same principle — the answer is built from live results and arrives with its list of sources.

How billing works

There is no subscription: you top up one balance and spend it on any model on the platform.

Pay per token

You are charged for exactly what the request and the answer used. Not using it costs nothing: no limits, no monthly fee.

One balance for every model

Perplexity, ChatGPT, Claude, image and video generation — all out of the same wallet. No separate subscription per service.

Top up the way you prefer

International cards, Apple Pay, Google Pay or crypto. The minimum top-up is $5.

See pricing

How to start with Perplexity

One account for every model on the platform.

Emailyou@example.comPassword••••••••

What has changed in advertising disclosure rules over the past six months and who do they apply to

Live search

Sonar Pro

What changed in the rules

The key change is a wider definition of what counts as an advertising creative, plus shorter deadlines for filing the details.

The duty now sits with the advertiser and the distributor at the same time, so «who files» can no longer be agreed after the fact.

Your own material about your own products still falls under the exemption — provided it carries no signs of promoting someone else's.

The model switches right inside the input — you can change it mid-conversation.

The answer arrives in the chat and the context is kept until the conversation ends.

Try Perplexity in GPTunneL

Signing up takes a minute, and $5 on the balance is enough to see whether the model fits your task.

Try Perplexity

Frequently asked questions

/01Do I need a VPN to use Perplexity through GPTunneL?

No. Requests go through GPTunneL's infrastructure, so the models open like any ordinary website — no VPN and no proxy.

/02Which Perplexity models are available in GPTunneL?

/03How is Sonar different from a regular model with web search?

/04What is Deep Research?

/05Can I trust the links in the answer?

/06Which languages does Perplexity handle?

Perplexity Sonar is a family of models from the American company Perplexity, built differently from an ordinary neural network. Sonar does not rely on what it memorised during training: for every question it goes to the web, builds the answer from live results, and attaches the list of sources it used.

That is why the family gets picked wherever currency and verifiability matter: breaking news, prices and rates, changes in rules and regulations, market overviews. A separate Deep Research model works a topic over many steps — it writes its own sub-queries, visits dozens of pages and assembles a full review.

Perplexity models are available in GPTunneL — worldwide, without a VPN, with one balance and billing for the tokens you actually spend.