News aggregation feeds
Since Jul 8Updated Nov 15, 2026Based on 110 AI answers
Reviewed by Dimitry Apollonsky
Brands AI mentions
A simpler, lower-cost API with predefined categories and AI-generated tags
Clean REST-oriented JSON and straightforward developer onboarding for market news
Turnkey AI-driven sentiment scores for news analysis
Also mentioned rarely: NewsAPI.ai, Perigon, Benzinga, GetStream, AYLIEN, Event Registry, GDELT, GNews, APITube, Mediastack, Polygon.io, Diffbot, EOD Historical Data, Massive, Firecrawl, The Guardian Open Platform, RSS.app, Tiingo, Currents API, Exa, Feedly, Webz.io, Contify, AP Metadata Services, Financial Modeling Prep, GDELT Cloud, Opoint, Quantexa, Benzinga APIs, Contify News API, RavenPack, finlight, Associated Press, Bing News Search API, Company News API, Polygon, StockNewsAPI, Weavy, Amity, Apify, Bloomberg, Dataminr, Associated Press Developer, FactSet Developer Portal, Article API, Dow Jones, Elasticsearch, FactSet, Firebase, Fold News, Stream Activity Feed API, ProsusAI/finBERT, Google, IAB Tech Lab, IFTTT, Intrinio, Lseg, MirrorFly, NewsBox, NewsEdge, NewsMesh, Nexis Data+, Novu, OpenSearch, Pabbly Connect, Pipeless, PubNub, Quantexa News Intelligence, Supabase, The Guardian, Twelve Data, Twingly, World News API, You.com
- NewsAPIChatGPTGoogle AI Mode
- NewsDataChatGPTGoogle AI Mode
- FinnhubChatGPTGoogle AI Mode
- NewsCatcherAPIChatGPTGoogle AI Mode
- Alpha VantageChatGPTGoogle AI Mode
Questions and answers
ChatGPTOct 25, 2026
If you want to add a news feed to your application, the best feed-as-a-service API depends on whether you want to aggregate news from publishers, deliver a personalized feed to each user, or build a social-style timeline.
These are three different problems, and the best provider for one may be overkill for another. For a typical application that needs to display current news articles, I’d start by comparing dedicated news APIs with managed feed infrastructure.
My top recommendations
1. GNews API
(Best starting point for a news feed)
A straightforward REST API for fetching headlines and searching articles across more than 80,000 sources. It supports categories, languages, countries, keyword searches, and historical data.
- Pricing: Free for development; paid Essential plan at €49.99/month.
- Best for: A news tab, topical feed, or industry-specific news section.
- Watch out: The free tier has a 12-hour delay and is not permitted for production commercial use.
2. NewsData.io
A strong alternative if you want broad international coverage, streaming news, historical archives, and optional AI-generated summaries or sentiment analysis.
- Pricing: Free tier with 200 API credits per day; paid plans offer larger quotas and more features.
- Best for: News discovery, dashboards, and data-heavy applications.
- Watch out: Credits, article limits, freshness, and historical access vary by plan.
Website ·
3. NewsAPI.org
A mature, developer-friendly option with a simple interface for searching articles and fetching top headlines from a large source index.
- Pricing: Free development tier; Business is listed at $449/month.
- Best for: Prototypes and applications that need a familiar, general-purpose news API.
- Watch out: The free plan is delayed and restricted to development and testing, and article text is limited.
4. The Guardian Open Platform
Worth considering if a single high-quality publisher is enough. It offers searchable Guardian articles and structured content through an API.
- Pricing: Free access options are available, subject to the current terms and use case.
- Best for: Editorial products, curated reading lists, and Guardian-focused feeds.
- Watch out: It does not aggregate news from multiple publishers.
What if you mean a true “Feed-as-a-Service” platform?
If you want the provider to manage feed infrastructure—personalized timelines, following topics or publishers, ranking, pagination, and real-time updates—rather than simply return news articles, evaluate a different category.
- Pipeless Activity Feeds — provides APIs for following feeds, activity timelines, and recommendations. Explore Pipeless.
- For a custom-built feed, you can also ingest news from a news API and manage ranking, deduplication, and user preferences in your own backend.
How I'd choose
Your requirement My pick Add a general news feed quickly GNews International news and enrichment NewsData.io Simple, established news API NewsAPI.org Personalized, activity-style timelines Pipeless Curated content from one publisher The Guardian API One important caveat: news-data licensing is often the deciding factor. Before launching, confirm your rights to display headlines, excerpts, article images, and full text, along with caching and attribution requirements.
One question to narrow it down
What kind of feed are you building?
1. What should the feed do?
Display news articles from many publishers
Personalize news for each user based on interests
Provide a social-style feed or timeline
Aggregate a niche, such as AI, finance, or industry news
2. What's your budget?
Free or open-source preferred
Under $100/month
$100–$500/month
Enterprise pricing is fine
3. What are you building for?
Web app
Mobile app
Both web and mobile
Backend service or AI application
[Find my best-fit API]
Pages cited in this answer13
- docs.gnews.ioGet Started with GNews API - GNews API Documentation
- gnews.ioPricing - GNews API
- gnews.ioTerms of Service - GNews API
- newsdata.ioAbout NewsData.io API | Documentation
- godofprompt.aiBest News API for Developers in 2026: 7 APIs Compared
- newsdata.ioBest News API for Real-Time & Historical News | NewsData.io
- newsapi.orgPricing - News API
- apiscout.devBest News APIs for Developers 2026 | APIScout
- apidirectory.devNews & Content APIs: 46 verified records · apidirectory
- itechguides.com6 Best News Scraper APIs and Tools for Developers in 2026
- newsapi.orgDocumentation - News API
- open-platform.theguardian.comOpen Platform
- pipeless.ioActivity Feeds API | Pipeless
Google AI ModeOct 25, 2026
When choosing a feed-as-a-service API , the "best" option depends entirely on whether you mean an in-app user activity/social feed (like a customized timeline, Reddit/Twitter style) or an external aggregated news feed (pulling global journalism or articles into your app).
Category 1: In-App & Social Feeds (User-generated or custom timelines)
If you are building a feed where your own users post, follow each other, like, comment, or share content:
- Stream Activity Feed API
- Best For: Scale, rich UI component integration, and real-time updates.
- Why it’s top-tier: Stream is the undisputed industry leader for feed-as-a-service. It handles complex architectural challenges like "fan-out" (push/pull architectures) effortlessly, scales to billions of users, and provides robust frontend SDKs (React, React Native, iOS, Android, Flutter) alongside native support for reactions, comments, and custom algorithmic ranking.
- Weavy
- Best For: B2B SaaS applications needing drop-in collaboration feeds and UI components.
- Why it’s top-tier: If you want a fast, secure, multi-tenant workspace feed (activity streams, comments, file sharing) without building a heavy social graph from scratch, Weavy offers seamless plug-and-play UI integration.
Category 2: External/Global News Aggregation Feeds (Third-party articles)
If you need an API that pulls real-time world news, global press, or specific niche topics into your application's feed:
- NewsData.io
- Best For: Multilingual reach, startups, and robust free tiers.
- Why it’s top-tier: Indexes over 100,000 sources across 200+ countries in nearly 90 languages. It is developer-friendly, includes AI-powered features/filtering, and offers a usable free tier for MVPs before scaling up to commercial plans.
- Perigon
- Best For: Enterprise media monitoring, event clustering, and deep analytics.
- Why it’s top-tier: Rather than just serving raw, chaotic lists of articles, Perigon excels at AI-driven event clustering—grouping hundreds of varying publisher perspectives on the exact same story into a clean, unified feed.
- APITube
- Best For: Sentiment and entity-aware news features.
- Why it’s top-tier: Unlike legacy competitors that hide sentiment analysis and entity extraction behind expensive custom enterprise contracts, APITube exposes these metrics right out of the box on standard tiers.
To help narrow down the ideal choice, tell me:
- Are you building a social/user-generated feed (like a community or timeline) or aggregating external news/articles?
- What is your expected monthly active user (MAU) count or API request volume?
Google AI ModeSep 26, 2026
For real-time filtering of news by specific entities (people, organizations, locations) and granular topics, traditional search endpoints fall short. You need News Intelligence Platforms that process high-throughput ingestion pipelines equipped with advanced Named Entity Recognition (NER), cross-lingual concept IDs, and hierarchical taxonomies.
The top vendors providing the most granular metadata tagging and filtering capabilities in their API responses include:
- NewsAPI.ai (formerly Event Registry)
- Granular Tagging & Metadata: Exceptional depth in NLP enrichment. It doesn't just extract entities (people, organizations, locations) with disambiguation; it links them to unique concept IDs and cross-language identifiers. It maps content across over 5,000 distinct, hierarchical topic categories.
- Real-Time & Events: Uniquely groups breaking news threads into dynamic "events" rather than just standalone articles, allowing you to track how a specific entity-driven story evolves.
- Explore further: Read the platform details on NewsAPI.ai.
- Perigon
- Granular Tagging & Metadata: Built explicitly as an enterprise news intelligence and semantic search engine. It maintains massive structured entity databases containing hundreds of thousands of pre-indexed people, companies, and journalists.
- Real-Time & Filtering: Provides precise query parameters and nested taxonomies for filtering articles or clustered stories by exact topic paths, entity sentiment, and source rank.
- Explore further: Review technical scopes via the Perigon Documentation and Overview.
- Webz.io
- Granular Tagging & Metadata: Offers deep contextual metadata tagging powered by advanced AI and LLM pipelines. It automatically extracts persons, organizations, and places while adding clean noise-reduction and sentiment metrics.
- Real-Time & Filtering: Excels at surfacing niche or first-party data (governmental/corporate feeds) pre-categorized into granular custom or standard topical clusters.
- Explore further: Check out the feature breakdown on Webz.io News API.
- Associated Press (AP) Metadata Services
- Granular Tagging & Metadata: If your workflow requires gold-standard industry taxonomies rather than purely automated AI extraction, the AP Metadata Services auto-classification system enriches content directly with standardized news vocabularies and mappings to the IAB Tech Lab Content Taxonomy.
If you'd like to narrow this down, let me know:
- Volume requirements (approximate number of requests/articles per day)
- Whether you need cross-lingual support or English-only
- Budget preference (self-serve vs. enterprise contract)
I can help you evaluate which of these APIs fits your exact architecture.
Pages cited in this answer10
- newsapi.aiBest News API 2025: 8 Providers Compared & Ranked
- newsdata.ioNews Categorization And Tagging In News API
- firecrawl.devBest News API for Apps and Agents in 2026
- newsapi.aiNews Aggregation: Build Smarter Platforms with Newsapi.ai
- mcpbundles.comPerigon MCP Server & Skill
- apitube.ioAPITube vs Perigon — News API Comparison | APITube News API - Access Real-Time News Worldwide
- sheetxai.comImport the Full Perigon Topic Taxonomy Into a Google Sheet | SheetXAI
- webz.ioNews API to Search, Monitor & Analyze Global News | Webz.io
- webz.ioWhat is a News API? How it Works?
- developer.ap.orgAP Metadata Services | AP Developer
ChatGPTSep 26, 2026
If your requirement is real-time filtering by specific entities + fine-grained topics, the vendors differ substantially. Based on their current API/documentation, I’d group them like this:
Vendor Entity metadata Topic taxonomy Real-time/event stream Granularity GDELT People, organizations, locations, names Thousands of themes + tone/emotion and other GKG fields ~15-minute global processing Very high Event Registry / NewsAPI.ai Disambiguated concepts/entities Categories, concepts, topics, events Minute-level streams Very high Quantexa / AYLIEN News API Entity extraction + relationships/metadata IPTC, IAB, Smart Tagger, industries Near-real-time news feed Very high NewsAPI.org Limited Broad publisher/category metadata Headlines are current, but enrichment is limited Low–medium 1. GDELT — deepest open metadata
GDELT Project is particularly strong if you want to build your own filtering layer.
Its Global Knowledge Graph extracts people, organizations, locations, themes, counts, and emotions from worldwide news, processing monitored news roughly every 15 minutes. Its query system can filter directly against themes and person/organization names rather than merely doing keyword/full-text matching.
That makes it unusually good for queries like:
PERSON = "X"ORGANIZATION = "Y"THEME = "FOOD_SECURITY"LOCATION = "Georgia"- combinations of entity/theme/location
The important caveat is that GDELT is a knowledge graph/news-monitoring system, not simply a conventional commercial article API. You'll generally do more normalization yourself.
2. Event Registry / NewsAPI.ai — probably the strongest fit for application-level filtering
Event Registry (the NewsAPI.ai API) exposes concepts, categories, events, sources, locations, authors, sentiment, etc. Concepts are entities/things that the system recognizes and disambiguates, rather than just strings appearing in article text.
It also has a particularly useful minute-stream capability. For example, its documentation shows filtering newly updated events by both a Tesla concept and a business category, with updates available over a minute-scale window.
For a product where your users say:
"Show me new stories about these 200 companies, but only when they're about M&A, litigation, earnings, or regulatory activity." this is very close to the API abstraction I'd want.
3. AYLIEN / Quantexa — strongest taxonomy-oriented option
The AYLIEN News API is now part of Quantexa. Its enrichment stack includes entity extraction, categorization, clustering, sentiment, industries and other NLP metadata.
Its Smart Tagger documentation is particularly notable: it describes 2,763 Smart Tagger categories, up to six hierarchical levels, alongside 1,403 IPTC News Codes and 404 IAB QAG categories. It also describes 26 NLP-derived data points per article.
That makes it attractive when "topic" means something more structured than a handful of generic categories.
4. NewsAPI.org — convenient, but not comparable on enrichment
NewsAPI is useful for straightforward article discovery and current headlines, but its metadata is substantially less granular. Its documented source metadata includes things such as publisher category, language and country, while its main endpoints are essentially article search/headline retrieval.
I wouldn't choose it as the primary enrichment layer if entity disambiguation + topic classification is the central requirement.
My practical shortlist
If I were designing this system, I'd evaluate these three first:
- Event Registry / NewsAPI.ai — best match for entity-aware, topic-aware real-time application filtering.
- GDELT — best for breadth, open access, extremely rich knowledge-graph metadata, and custom analytics.
- AYLIEN/Quantexa — best when taxonomy depth and commercial NLP enrichment matter more than having an open ecosystem.
One distinction is especially important: "metadata richness" and "filterability" aren't identical. GDELT can expose an enormous amount of structured information, whereas Event Registry's API abstractions can be easier to turn directly into user-facing saved filters and alert rules.
If you tell me your required entities/topics (e.g. companies + people + products + geopolitical topics + events) and your latency target (e.g. <1 min, <5 min, <15 min), I can make a more technical comparison of the APIs, including actual response schemas, taxonomy depth, entity IDs/disambiguation, webhook/stream support, coverage, pricing, and rate limits.
Pages cited in this answer6
- NewsAPI.ai (formerly Event Registry)
Google AI ModeSep 27, 2026
For a developer-centric financial sentiment analysis tool, you need clean JSON payloads, precise entity-to-ticker mapping, and reliable uptime. Several developer-friendly vendors stand out for this exact use case, balancing pre-calculated sentiment scores with raw article ingestion.
- Marketaux API is purpose-built for clean, structured JSON payloads featuring deep entity recognition. It automatically maps articles to specific stock tickers, indices, crypto, or forex, and delivers pre-calculated, entity-level sentiment scores (ranging from -1 to 1) for each company mentioned inside a single article. It includes a generous free tier for development.
- Finnhub Stock APIs provides a dedicated News Sentiment Endpoint alongside general company news. Finnhub aggregates multi-source feeds and returns aggregated sentiment stats, buzz scores, and regional daily sentiment changes alongside raw article JSON, making it easy to build dashboards or algorithmic triggers.
- Alpha Vantage News API integrates live global market news directly with built-in AI-derived sentiment scores. Querying by ticker or topic yields structured JSON containing relevance scores and explicit sentiment labels (e.g., Somewhat-Bullish, Bearish ) for each article, minimizing the need to run an external NLP pipeline from scratch.
- EODHD Financial APIs offers a robust financial news feed endpoint paired with built-in sentiment calculations. It continuously processes major global portals for stocks, ETFs, and crypto, returning uniform JSON payloads containing positive/negative mention metrics and daily aggregated sentiment scores via standard REST or SDK options.
- Benzinga APIs is an industry favorite if your tool relies heavily on fast, market-moving U.S. equity headlines, earnings whispers, and corporate actions. While more premium-focused, their developer portal delivers high-velocity, clean JSON streams ideal for real-time trading or immediate sentiment reaction tracking.
If you can share what asset classes you are focusing on (crypto, US equities, or global forex ) and whether you plan to rely entirely on vendor-provided sentiment or pipe raw text into your own custom LLM/NLP model , I can help you pick the most cost-effective provider.
Pages cited in this answer6
- publicapis.devMarketAux: Live stock market news with tagged tickers - Public APIs
- eodhd.comFinancial News Feed and Stock News Sentiment data API | EODHD APIs Documentation
- qveris.aiFinancial News APIs for AI Agents: 6 Providers Compared | QVeris Guides
- github.comGitHub - api-evangelist/marketaux: MarketAux — independent third-party profile of a public API surface, by API Evangelist. Marketaux is a global financial and stock market news API that pairs every article with entity recognition and sentiment analysis. It aggregates 5,000+ news sources in 30+ languages, tracking 200,000+ entities — equities, indices, ETFs, mutual funds, currencies, · GitHub
- finnhub.ioNews Sentiment
- alphavantage.coFree Stock APIs in JSON & Excel | Alpha Vantage
ChatGPTSep 27, 2026
For a developer-centric financial sentiment pipeline, I’d shortlist vendors based on four things: real-time delivery, structured JSON, ticker/entity mapping, and whether you can legally process/store the content for your model.
Vendor Best fit Real-time JSON/API Sentiment Key trade-off finnhub.io Fastest MVP / broad financial API WebSocket + REST Yes Built-in news sentiment Premium access for real-time news benzinga.com Production-grade market-news NLP REST + TCP streaming + RSS Yes You analyze it yourself More enterprise-oriented/pricing is quote-based alphavantage.co Simple API + sentiment prototype Near-real-time/latest news Yes Built-in sentiment Less compelling if you need high-volume streaming polygon.io News tightly integrated with market data Streaming/REST ecosystem Yes You analyze it yourself News depth/licensing needs closer evaluation for your use case 1. Finnhub — probably the easiest starting point
Finnhub is unusually convenient if your application needs both news and market data from one API. Its REST API returns JSON, and its premium WebSocket feed can stream real-time news for US and Canadian stocks. News records include timestamp, headline, source, summary, URL, and related securities.
It also exposes a news-sentiment endpoint, including bullish/bearish percentages and a company news score.
That makes it particularly useful for building a prototype like:
Finnhub WebSocket ↓ news event ↓ ticker/entity normalization ↓ your sentiment model ↓ Kafka / Redis / database ↓ signals + dashboard2. Benzinga — strongest candidate for a serious news-first system
benzinga.com is more specifically focused on market-moving financial news rather than being a general-purpose financial-data API.
It offers REST, TCP streaming, and RSS, with JSON support, US-equity/macroeconomic coverage, and historical news going back to 2010. Benzinga says its newswire content is produced in-house rather than scraped/aggregated.
That's attractive for sentiment analysis because you generally want:
- consistent timestamps
- clean headlines/text
- explicit financial relevance
- reliable ticker associations
- low duplication
- predictable delivery latency
Benzinga also advertises sub-0.1-second API latency and a large historical archive across its broader API offering.
3. Alpha Vantage — attractive for a sentiment MVP
alphavantage.co is worth considering if you specifically want news + precomputed sentiment in one API and don't initially need a sophisticated streaming architecture.
Its big advantage is simplicity: you can retrieve financial news and sentiment metadata without having to build the entire NLP layer yourself. For an experimental system, that's useful as a benchmark against your own model.
I'd treat it more as an MVP/research option than automatically assuming it's the right source for a high-throughput, latency-sensitive production feed.
4. Polygon — worth evaluating if news is only one component
polygon.io makes more sense when your system also needs a substantial market-data stack. If your eventual architecture correlates news sentiment with ticks, trades, quotes, aggregates, etc., having those datasets under one API ecosystem can simplify engineering.
What I'd choose
For your particular description, I'd evaluate them in this order:
- Benzinga — if news quality, latency and production NLP are the core product.
- Finnhub — if you want the quickest path to a working developer API combining news, securities and sentiment.
- Polygon — if news is part of a broader real-time market-data infrastructure.
- Alpha Vantage — if you're primarily prototyping/researching sentiment.
One important caveat: "real-time API" doesn't necessarily mean you can freely feed the returned article text into a commercial ML system or retain it indefinitely. Before choosing, compare redistribution, storage, derived-data, model-training, and historical-access rights—not just API latency and price.
If you tell me your expected events/sec, US-only vs global coverage, budget, and whether you need full article text or just headlines/summaries, I can narrow this to a concrete 2–3 vendor architecture and compare their APIs/schema/latency/licensing.
Pages cited in this answer6
- finnhub.ioFinnhub Stock APIs - Real-time stock prices, Company fundamentals, Estimates, and Alternative data.
- benzinga.comFinancial Market Data & APIs - Benzinga
- alphavantage.coFree Stock APIs in JSON & Excel | Alpha Vantage
- polygon.ioStock Market API | Massive
- finnhub.ioGlobal Stocks, Forex, Crypto price
- benzinga.comStock Market Newswire - Benzinga APIs
Also asked as
- What are the top-rated API-first solutions for embedding a news feed into an application?
- What are the best real-time news APIs for a developer building a financial sentiment analysis engine that requires clean JSON integration?
For agents: this page as Markdown, https://parse.gl/topics/news-aggregation-feeds.md. JSON, MCP · free to read, $0.01 a check
This page
Ask and search
{"measures": ["share_named"], "by": ["brand"], "filter": {"topic": ["top_e6d8525a-a953-4b16-9653-253729c0041f"]}}Connect
An agent can start a workspace with no account: POST https://agent.parse.gl/v1/workspaces
What it costs
Prices: https://parse.gl/pricing.md
Map of the site: https://parse.gl/llms.txt