Perplexity brings live market data to its Agent API

If you have ever tried to build an AI agent for finance, you already know the nightmare. You have to wire up one API for real-time stock ticks, another for balance sheets, and a web scraper for earnings call transcripts. It is messy, expensive, and prone to hallucinations.

Perplexity wants to fix that. The company just rolled out Finance Search inside its Agent API, packing licensed financial datasets, real-time market data, and cited web sources into a single tool call.

Direct access replaces generic web crawling

Instead of making an LLM crawl messy web search results to find a company's revenue, Finance Search queries structured, licensed financial databases directly.

Developers can pull valuation metrics, balance sheets, cash flow statements, earnings call transcripts, and analyst estimate revisions with one standardized request.

Every data point comes back in a consistent schema with inline citations, showing exactly which provider produced each specific figure.

On the FinSearchComp T1 benchmark, tested right after market close, Perplexity claims Finance Search achieved the highest retrieval accuracy and the lowest cost per correct answer, simply because models process fewer irrelevant tokens.

Perplexity is already dogfooding the feature in its own Computer for Professional Finance product to generate research memos and tearsheets.

Why it matters

In finance, being almost right is the same as being wrong. A hallucinated decimal point or a stale quarterly figure ruins an entire valuation model.

For developers building financial agents, this removes months of boring infrastructure work. Instead of negotiating data contracts with multiple financial vendors and parsing incompatible payloads, you get a clean endpoint with built-in audit trails. If you are building automated research or market monitoring tools, it is definitely worth testing.

It is also a shrewd strategic move for Perplexity. While generic search engines are busy summarizing blog posts, Perplexity is moving up the value chain directly into enterprise territory where high-margin financial tools live.

That said, we think bundling all your financial feeds behind one third-party wrapper creates obvious vendor lock-in. If Perplexity loses a data license or changes its pricing model down the road, your entire pipeline is at the mercy of their API.