---
type: "post"
id: "exa-launched-exa-for-finance-2b-financial-documents-63edf8"
slug: "exa-launched-exa-for-finance-2b-financial-documents-63edf8"
title: "Exa launched Exa for Finance, 2B+ financial documents searchable through the existing API"
description: "Exa’s Financial Markets documentation describes search for market data, filings, earnings-call transcripts and economic releases through Exa Search."
retrieval_nugget: "Exa’s Financial Markets documentation describes search for market data, filings, earnings-call transcripts and economic releases through Exa Search. For teams building financial research agents, the significance is a documented way to retrieve several kinds of financial information with natural-language queries, while structured multi-source work can be handed to Exa Agent."
published_at: "2026-11-10"
updated_at: "2026-11-10"
record_date: "2026-11-10"
date_kind: "published_at"
topics: ["retrieval"]
entities: ["Exa","Exa Search","Exa Agent","Parallel"]
related_patterns: []
novelty_label: "new_fact"
what_changed: "Exa launched a finance-focused search offering covering more than 2 billion financial documents through its existing API."
prior_refs: ["https://newruntime.com/posts/parallel-for-finance-web-search-for-financial-ai-ae1a17/"]
source_url: "https://exa.ai/docs/search/data/financial"
source_urls: ["https://exa.ai/docs/search/data/financial"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:exa-launched-exa-for-finance-2b-financial-documents-63edf8"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/exa-launched-exa-for-finance-2b-financial-documents-63edf8/","markdown":"https://newruntime.com/posts/exa-launched-exa-for-finance-2b-financial-documents-63edf8.md","json":"https://newruntime.com/posts/exa-launched-exa-for-finance-2b-financial-documents-63edf8.json"}
markdown_url: "https://newruntime.com/posts/exa-launched-exa-for-finance-2b-financial-documents-63edf8.md"
---

# Exa launched Exa for Finance, 2B+ financial documents searchable through the existing API

## Retrieval answer

Exa’s Financial Markets documentation describes search for market data, filings, earnings-call transcripts and economic releases through Exa Search. For teams building financial research agents, the significance is a documented way to retrieve several kinds of financial information with natural-language queries, while structured multi-source work can be handed to Exa Agent.

## Summary

Exa’s Financial Markets documentation describes search for market data, filings, earnings-call transcripts and economic releases through Exa Search. For teams building financial research agents, the significance is a documented way to retrieve several kinds of financial information with natural-language queries, while structured multi-source work can be handed to Exa Agent.

## What happened

Exa’s documentation presents Financial Markets as a data category for Exa Search. It says the search surface can find prices, filings, transcripts and reporting around them in one query. Its example is a question about a ticker that can return a quote, the latest earnings call and analyst coverage together. The listed data includes quotes and recent price history for stocks, crypto, foreign exchange, indices, futures, options and commodities; security profiles with key statistics and daily OHLCV history; and earnings-call transcripts with prepared remarks and question-and-answer sections attributed to speakers.

The documentation also lists SEC filings, reported financials and international filings, alongside analyst estimates, funding announcements and economic releases. It describes potential applications including equity and credit research, KYC and KYB screening, adverse-media screening, portfolio and policy monitoring, deal sourcing and private-market research. These are use cases Exa proposes; the page does not establish that a particular team has deployed them or achieved a measured result.

Its examples show how a query can specify the task rather than just request a broad search. To retrieve a transcript, Exa recommends naming the company and quarter; to find a disclosure, the example asks for 10-K risk factors mentioning dependency on third-party AI models and sets the search category to “financial report.” A private-market query can specify the round, sector and time window, while an economic-data request can name the release and the figure sought. The page includes Python, JavaScript and cURL examples using Exa’s search endpoint, with `type` set to `auto`, a result count, and a financial-report category in the filing example.

For work that requires structured data gathered across multiple sources, Exa directs users to an Exa Agent task run. The documentation says a user can describe securities, periods, criteria and desired output fields; Agent returns schema-validated results with citations. It points readers to the Agent guide for screening companies, comparing filings or assembling a structured portfolio brief. The page does not give a price, a date for the feature, or quantified accuracy or coverage guarantees in the described material.

## What the archive adds

New Runtime’s earlier coverage on October 30, 2026 described Parallel’s finance-specific index and ranking upgrade across its APIs, including reported benchmark accuracy and cost advantages without a new subscription ([Parallel for Finance: web search for financial AI](https://newruntime.com/posts/parallel-for-finance-web-search-for-financial-ai-ae1a17/)). Exa’s documentation is a distinct product surface for financial search, not evidence of a change to Parallel’s offering. This is a new fact in the broader move to finance-oriented retrieval, rather than a restatement of the earlier event.

## Trend signal

This does not yet fit a pattern New Runtime tracks. Exa documents financial search capabilities and an agent workflow, but the page alone does not establish a broader shift in interoperability, runtime security or verified task economics.

## Connections to other events

The connection to New Runtime’s October 30, 2026 coverage of Parallel for Finance is specific: both offerings apply specialized retrieval to financial research. Parallel’s earlier coverage concerned a finance-specific index and ranking upgrade across its APIs; Exa’s documentation describes its own searchable financial data and an option for schema-validated, cited results through Agent. These are adjacent approaches, not evidence that the products share an index or evaluation method.

## Why it matters

Financial research agents are only as useful as the evidence they can retrieve and the provenance they return. Exa’s examples make the query design practical: ask for a particular filing disclosure, named company and reporting period, or an economic release and value. A financial product team can test whether a search API returns the primary document or transcript it needs, rather than relying on a model’s stored knowledge or a general web result.

The distinction between search and Exa Agent also matters architecturally. Search is presented for targeted retrieval; Agent is presented for research across sources with a schema and citations. That gives builders a possible division of labor, but not a guarantee that output is complete, current or fit for regulated decisions. Teams should evaluate retrieval against their own instruments, periods and source requirements, and preserve the underlying documents for human review where decisions carry financial or compliance consequences.

## Limits and uncertainty

Exa’s page describes product capabilities and example queries, not an independent evaluation. It does not establish recall or precision across asset classes, freshness and update frequency for each data type, geographic coverage, or how conflicts between sources are resolved. It also does not quantify the availability of individual filings, transcripts or historical series. A schema-validated result with citations is not itself proof that the underlying extraction is correct. Teams should verify relevant documents and test coverage, latency, and failure behavior for the markets and workflows they intend to support.

## Skill to build now

Build a small retrieval test set from the financial questions your team actually handles: a quote and date, an earnings-call passage, a filing disclosure, and a dated economic release. For each case, record the expected primary source and fields, then compare Exa Search’s returned documents and Exa Agent’s cited structured output against that reference before connecting either to consequential decisions.

## Reader takeaway

Exa documents a finance-specific search surface that can bring market data, filings, transcripts and economic releases into agent workflows. Treat it as a retrieval capability to test, not a quality guarantee: verify source coverage and citations against the financial documents your application must get right.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Retrieval](https://newruntime.com/topics/retrieval.md): Explore the retrieval topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Sources](https://newruntime.com/sources/): See the original sources this archive cites.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)