Analysis Summary
Perplexity Sonar is a text-and-image model released in January 2025, with a 127K context window. Its measured general capability is limited, while its image input and substantial context offer practical flexibility for interpreting visual material alongside text. The available coding result is modest, and the supplied record does not indicate tool use or function calling.
It may suit lightweight content assistance, image-aware text tasks, or summarization where a large input window is useful. For agency adoption, the main constraint is a lack of listed workflow integration features and limited evidence of coding or high-reliability reasoning. The listed pricing is $1 per million input tokens and $1 per million output tokens, so it is not the cheapest option for routine high-volume generation.
Consider Sonar for a focused multimodal pilot rather than as a general-purpose automation engine. Validate factual accuracy and instruction following on representative client material, and compare the benefits of image input with the available cost and workflow needs.
Assessed September 23, 2026
Editorial notes
Perplexity Sonar pairs a 127K context window with image input and measured general capability. Its coding results are limited, and the supplied data does not list tool or function-calling support.
The DFO score ranks models on the Artificial Analysis Intelligence Index and is refreshed as new models launch. The write-up and verdict are our own. Feedback?
DFO Verdict
Perplexity Sonar pairs a 127K context window with image input and measured general capability. Its coding results are limited, and the supplied data does not list tool or function-calling support.
Benchmarks
Artificial Analysis data refreshed Nov 16, 2026. Arena ratings from the Nov 9, 2026 text leaderboard, with style control. Bars: magenta = reasoning, ink = coding and agents, cyan = instructions and long context.
Artificial Analysis has not published coding or agentic results for Perplexity: Sonar yet. Until it does, the model sits below tested models on the Coding and AI Agents lists.
Artificial Analysis no longer lists this model, so its indexes are converted from their older scale to stay comparable.
How Perplexity: Sonar compares
Perplexity: Sonar ranks #271 of 454 AI models we track for overall intelligence. Its 127K-token context window is larger than 29% of the models we list. At $1.00 per million input tokens it is cheaper than 25% of comparable models.
Dark bar = input · light bar = output, scaled to the priciest peer.
1M tokens ≈ 8 full-length novels or ~2,500 pages of business documents in a single request.
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $1.00 | $0.001000 |
| Output | $1.00 | $0.001000 |
What would Perplexity: Sonar cost your business?
Pick the job that looks most like yours, then fine-tune with the sliders. Estimates update live.
A website chatbot handling around 100 customer conversations a day, a few short messages each.
Full calculator with 749 models → Price Calculator
These numbers get smaller with the right architecture.
We route routine calls to cheap models and save Perplexity: Sonar for the hard ones. Most clients cut their estimate by 60-80%.
Talk to our teamAbout Perplexity: Sonar
Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources. It is designed for companies seeking to integrate lightweight question-and-answer features..
Explore Related Models
Frequently asked questions about Perplexity: Sonar
How much does Perplexity: Sonar cost?
Perplexity: Sonar costs $1.00 per million input tokens and $1.00 per million output tokens.
What is the context window of Perplexity: Sonar?
Perplexity: Sonar has a context window of 127,072 tokens (127K).
What can Perplexity: Sonar do?
Perplexity: Sonar supports image/vision input.
Who created Perplexity: Sonar?
Perplexity: Sonar is developed by Perplexity and was released on January 27, 2025.
Benchmarks from Artificial Analysis, pricing from the OpenRouter API, and verdicts from our own testing.
Last updated: November 16, 2026 7:38 pm