Back to models
Chat Models

Perplexity

Sonar Pro

Perplexity’s Sonar Pro supports deeper search and a 200K-token context. Explore its reference pricing and Agent API transition.

CitationsReasoning

At a glance

Know the model before you prompt.

Perplexity API specifications
Context window
200,000 tokens
Maximum output
Not verified
Inputs → output
Text → Text
Knowledge cutoff
Not verified

API model ID: sonar-pro

Capabilities & boundaries

What it supports. Where the limits are.

Tool support requires the appropriate API integration; a supported tool is not automatically active in every chat.

Supported API features and tools

  • Advanced web retrieval for complex queries
  • A documented 200K-token context
  • Citations and search-result metadata
  • Streaming and follow-up search workflows

Before you choose

  • More retrieved results do not automatically produce a better-supported answer.
  • Base Sonar Pro, Pro Search, and Sonar Reasoning Pro are distinct offerings.
  • Published token prices do not include every search-mode or request charge.

Non-reasoning model

The base Sonar Pro model is documented as non-reasoning. Pro Search is a separate enhanced search mode; it is not the same thing as Sonar Reasoning Pro. Verify mode and integration support before relying on it.

  • Perplexity’s migration guide maps Sonar Pro to the Agent API fast preset. Presets and underlying model selections can change; test results after migrating.
  • The reviewed model card does not publish a maximum output-token limit or training cutoff. Web access does not make an unknown training cutoff a current date.
  • This guide’s examples use text. Check the selected endpoint’s current input and output support rather than assuming full parity with another Sonar or Agent API configuration.

Put it to work

Start with a more useful prompt.

Original examples from EZ Ai Assist. Adapt these to your task and the features available in your workspace.

Workflow 01

Compare vendor documentation

Make a source-backed comparison without relying on promotional summaries.

Compare the three vendors below using only current official documentation. Build a table of supported deployment options, integration methods, published limits, and pricing-page links. Cite every populated cell. Use not published when a detail is missing and include an as-of date for the comparison.

Workflow 02

Track a release across sources

Keep version-specific changes separate from general positioning.

Research the latest stable release of this developer tool. Reconcile its release notes, migration guide, and official issue tracker into a brief covering breaking changes, upgrade steps, and unresolved caveats. Cite each claim and flag conflicts between sources instead of silently resolving them.

Workflow 03

Build a topic reading brief

Use multiple sources with a clear editorial boundary.

Create a reading brief on the topic below using five credible sources, prioritizing primary documentation. For each source explain the question it answers, publication date, and one limitation. Finish with a short synthesis distinguishing established facts from open questions. Do not treat repeated claims as independent corroboration.

Developer reference

Perplexity API pricing

These are Perplexity Sonar API reference prices, not EZ Ai Assist subscription prices.

View EZ Ai Assist plans
Published Sonar reference · USD per 1,000,000 tokens
Token typePrice
Input$3.00
Output$15.00
Additional API fees · USD
UsageRate and unit
Low search context$6.00per 1,000 requests
Medium search context$10.00per 1,000 requests
High search context$14.00per 1,000 requests
  • These are the published Sonar reference rates, not an estimate of migrated Agent API costs or OpenRouter resale prices. Check the current endpoint, preset, and response usage before budgeting.
  • Add the applicable search-context request fee to token charges. Search context controls retrieved material, not reasoning effort.
  • Pro Search has separate pricing considerations; the table describes base Sonar Pro.
  • Perplexity recommends Agent API for new projects. Migration can change both execution and billing; retain these rates as a clearly scoped Sonar reference.

Common questions

A few things worth knowing.

When should I consider Sonar Pro?

Compare Sonar Pro for broader searches, follow-up questions, and source synthesis that needs more context than standard Sonar. Its documented base model is non-reasoning; Perplexity’s separate Pro Search behavior and migration path should not be confused with that base model.

What changed with the Sonar API?

Perplexity says Sonar Chat Completions support ended September 27, 2026. Synchronous and streaming calls continue through a gradual translation to Agent API requests; async Sonar calls are no longer supported. This does not establish whether EZ Ai Assist or a reseller exposes a particular model.

Which Agent API preset is the documented replacement?

Perplexity maps Sonar Pro to fast. Treat this as migration guidance, not an identical model or a guaranteed equivalent response. Recheck citations, quality, latency, and usage on representative tasks.

Are search fees included in token prices?

No. The published Sonar table adds request fees based on search-context size. The low, medium, and high rows are per 1,000 requests, not per million tokens.

Is Sonar Pro a reasoning model?

The base model card describes Sonar Pro as non-reasoning. Pro Search is a separate mode, and Sonar Reasoning Pro is another model. A multi-step question does not make these names interchangeable.

What do the limits on this page mean?

The model card lists a 200K-token context. It does not establish a maximum output limit or training cutoff in the reviewed documentation. Integration limits may be lower, and search-backed answers still need checking.

Are these fees part of my EZ Ai Assist subscription?

No. The reference table describes Perplexity’s documented API billing. EZ Ai Assist plans are separate, and app availability must be checked in the workspace. These prompts are original starting points, not verified benchmark outputs.

Check the source

Official documentation

Specifications and API prices checked on . Example prompts and workflow advice are editorial guidance from EZ Ai Assist.

Same provider