Perplexity Launches a Context API That Reads Up to 10 URLs Per Call
Perplexity Context API is live. Context API reads up to 10 URLs per call and returns clean .md, HTML, links, and snippets per page. It returns usable content for 92.3% of URLs, 87.9% fact recall, and 80.3% content precision at $1/1000 URLs fetched.
Context
A priced URL reading API with the company's own quality scores is a practical tool for agents that need clean page text, but those accuracy figures are Perplexity's, not an outside test.
Perplexity Developers posted that the Perplexity Context API is live. The post said the API reads up to 10 URLs per call and returns clean Markdown, HTML, links, and snippets per page, and it reported usable content for 92.3% of URLs, 87.9% fact recall, and 80.3% content precision at $1 per 1,000 URLs fetched. Those quality and price figures are Perplexity's own. The post links a product page.
Deep dive
The official Perplexity Developers account said the Perplexity Context API is live. According to the post, the API reads up to 10 URLs per call and returns clean Markdown, HTML, links, and snippets per page. It also reported three quality figures, plus a price: usable content for 92.3% of URLs, 87.9% fact recall, and 80.3% content precision, at $1 per 1,000 URLs fetched. A video and a product page are attached.
Those are first hand product and pricing claims. The 92.3%, 87.9%, and 80.3% results are Perplexity's own measurements. The post does not describe the evaluation set, does not name a third party auditor, and does not compare the API with rival URL readers in the text. Those scores should stay attributed.
The practical offer is a priced, structured page reading interface for agents and apps, with a cap of 10 URLs per call. That is incremental tooling, useful if the cleanliness claims hold, and not a new frontier model.
Because the lead is official, the launch, the 10 URL cap, the return formats, the listed price, and the company's quality figures are confirmed as Perplexity's announcement. Independent quality is not confirmed here.
What to expect next
The product page should be checked for limits, authentication, and a methods note on the three quality scores. Watch early developer tests against messy pages, paywalls, and non English URLs, and whether the $1 per 1,000 price holds. If a later official post revises the scores or the cap, this story should be updated. The item should stay a tooling launch, not be folded into Perplexity's other Nov 17 API stories unless the company ties them together.
