Perplexity Sonar Pro: what it does and costs
perplexity/sonar-proSonar Pro is Perplexity’s text model for text and image inputs, with a 200,000-token context window.
200K tokens of context — about 150,000 words, on a scale that ends at 2M.
What it can do
- Vision
- Takes text
- Takes image
- Gives text
Released 7 Mar 2025 · checked 3 Oct 2026.
About it
Sonar Pro is a Perplexity model that accepts text and images and returns text. It has a 200,000-token context window and can produce up to 8,000 tokens in one answer. Its listed prices are $3 per million input tokens and $15 per million output tokens; those prices include Perplexity search pricing. It is available through provider APIs.
Review
In a line
Sonar Pro is a Perplexity model that takes text or image inputs and responds with text.
What it's good at
It suits questions that benefit from Perplexity search and more involved, multi-step queries. It can also accept images, alongside text.
Price in plain terms
Input costs $3 per million tokens, and output costs $15 per million tokens. The listed price includes Perplexity search pricing.
At those rates, reading a page of about 500 words costs about 0.19 cents. A chat turn with about 1,000 input tokens and 500 output tokens costs about $0.01; 1,000 such turns cost about $10.50.
Context and limits
Sonar Pro can take in up to 200,000 tokens at once and write up to 8,000 tokens in one answer. Its listed output is text.
Ways to use it
You can use Sonar Pro through Perplexity’s API or through providers that offer it. Browse Perplexity models to see related options.
Verdict
Consider Sonar Pro if you need a model that accepts both text and images, supports long inputs, and is suited to detailed, multi-step questions. Its input price is lower than its output price, so answers that generate a lot of text can cost more. Compare it with other vision models if image input matters to your choice.
Verdict
Sonar Pro may suit people who want text and image input, a large context window, and answers for detailed, multi-step queries. Its per-token costs are easy to estimate, though output tokens cost five times as much as input tokens. The listed rates include Perplexity search pricing, so they are not a separate search add-on. If long inputs are your main concern, compare the 200,000-token window with other models with long context windows before choosing.
- Accepts both text and image inputs.
- Can take up to 200,000 tokens in one prompt.
- Can produce up to 8,000 tokens in one answer.
- Prices include Perplexity search pricing.
- Output tokens cost $15 per million, more than input tokens.
- Its listed output format is text.
- 1,000 example chat turns cost about $10.50.
- Pick it ifPeople asking detailed, multi-step questions; Users who need image and text input
- Skip it ifYou need output other than text; You want lower costs for generating lots of text
Pricing
Input is $3 per million tokens and output is $15 per million tokens. The rates include Perplexity search pricing. A 500-word page costs about 0.19 cents to read; 1,000 example chat turns cost about $10.50.
Questions
Who makes Sonar Pro?
Perplexity makes Sonar Pro.
Can Sonar Pro take images?
Yes. It accepts both text and image inputs, and returns text.
How much text can it read at once?
Its context window is 200,000 tokens.
How long can one answer be?
It can produce up to 8,000 tokens in one answer.
What does Sonar Pro cost?
Input costs $3 per million tokens and output costs $15 per million tokens. These prices include Perplexity search pricing.
How much might everyday use cost?
A page of about 500 words costs about 0.19 cents to read. A sample chat turn costs about $0.01; 1,000 such turns cost about $10.50.
How can I use Sonar Pro?
It is available through Perplexity’s API and through providers that offer the model.