Perplexity on Tuesday launched an API service called Sonar, allowing enterprises and developers to build the startup’s generative AI search tools into their own applications.
“While most generative AI features today have answers informed only by training data, this limits their capabilities,” Perplexity wrote in a blog post. “To optimize for factuality and authority, APIs require a real-time connection to the Internet, with answers informed by trusted sources.”
To start, Perplexity is offering two tiers that developers can choose from: a base version that’s cheaper and faster, Sonar, and a pricier version that’s better for tough questions, Sonar Pro. Perplexity says the Sonar API also gives enterprises and developers the ability to customize the sources its AI search engine pulls from.
With the launch of its API, Perplexity is making its AI search engine available in more places than just its app and website. Perplexity says that Zoom, among other companies, is already using Sonar to power an AI assistant for its video conferencing platform. Sonar is allowing Zoom’s AI chatbot to give real-time answers, informed by web searches with citations, without requiring users to leave the video chat window.
Sonar could also give Perplexity another source of revenue, which could be particularly important to the startup’s investors. Perplexity only offers a subscription service for unlimited access to its AI search engine and some additional features. However, the tech industry has slashed prices to access AI tools via APIs in the last year, and Perplexity claims to be offering the cheapest AI search API on the market via Sonar.
The base version of Sonar offers a cheaper and quicker version of the company’s AI search tools. Sonar’s base version has flat pricing and uses a lightweight model. It costs $5 for every 1,000 searches, plus $1 for every 750,000 words you type into the AI model (roughly 1 million input tokens), and another $1 for every 750,000 words the model spits out (roughly 1 million output tokens).
The pricier Sonar Pro gives more-detailed answers and is capable of handling more-complex questions. This version will run multiple searches on top of a user prompt, meaning the pricing could be more unpredictable. Perplexity also says this version offers twice as many citations as the base version of Sonar. Sonar Pro costs $5 for every 1,000 searches, plus $3 for every 750,000 words you type into the AI model (roughly 1 million input tokens), and $15 for every 750,000 words the model spits out (roughly 1 million output tokens).
Perplexity claims Sonar Pro outperformed leading models from Google, OpenAI, and Anthropic on a benchmark that measures factual correctness in AI chatbot answers, SimpleQA.
As we recently reported, Perplexity’s annual recurring revenue is somewhere between $5 million and $10 million. This seems fairly healthy for a startup of Perplexity’s size and age, but the startup is certainly looking for new ways to grow its revenue. The startup raised an additional $73.6 million in a funding round earlier this month, valuing the company around $520 million.
