dxFeed review

dxFeed is a financial market-data service and analytics provider, not a broker. In a trading or research stack, it sits upstream of the application: it supplies market data that a platform can display, analyze or use in trading workflows. Its offerings include historical, real-time, delayed, calculated and aggregated data, as well as analytics applications and other data services.

The vendor says its historical service parses, indexes and stores up to 10Tb of raw data daily, with millisecond time-stamped tick data stored using cloud computing. Its Data Lake platform is aimed at institutional customers needing fast access to historical data, and can receive market data, corporate actions, fundamentals and metadata. For end users, dxAnalytics, widgets and Inforider provide web-based analysis and visualization. dxFeed also says its Chainlink oracle node makes financial market data available to blockchain applications. This is a set of hosted services and applications; the vendor does not describe a self-hosted server in its homepage text.

The vendor describes an adaptable data layer that integrates its widgets with various dxFeed APIs, but does not name a protocol, SDK, programming language or CLI on the homepage. A Reddit user reports connecting to a dxFeed token through a JavaScript WebSocket and subscribing to TradeEvents and CandleEvents. That is a user-reported integration path, not a published commitment to a particular client library or supported runtime.

The homepage presents web applications and embeddable widgets as integration options, alongside data services and APIs. For a production build, ask dxFeed which API and transport apply to your market, what authentication and subscription model is required, and which SDKs and languages are supported. Confirm whether the relevant data is available under your account entitlements and whether the widget data layer fits your application. Developers should also confirm event schemas, subscription error behavior, historical access methods and any platform-specific requirements in the vendor's documentation before choosing an integration.

The available pricing information does not list a starting price, named plans, add-ons, a free tier or a free trial. The reported pricing data says subscriptions are billed on a recurring 1-month/30-day basis and that annual billing is not listed, but this information has low confidence. Treat the cost as quote-based until dxFeed confirms it for your markets and use case.

That leaves key production cost drivers to establish directly with the vendor: which exchanges and instruments are included, whether real-time or delayed access changes the price, how historical data is priced, and whether display, redistribution or non-display use requires separate permissions. Ask about user or connection charges, minimum commitments, entitlement fees, usage caps and overages before estimating total cost. The homepage describes retail subscription services for dxFeed partners, but does not publish partner pricing. There is no published free allowance to use for development or staging, so request a test environment and written pricing terms before committing an application to the feed.

The vendor's homepage does not publish an open-source license or identify source code for the market-data service. It describes cloud computing for historical storage and a Data Lake platform, but does not state that customers can self-host dxFeed's infrastructure. Ask the vendor for deployment options, data residency terms and any licenses that apply to client libraries or integrations.

The homepage also does not describe a data-export process or a migration path. Before adopting dxFeed, confirm which data you can retain and export, in what formats, and whether your application can keep normalized records independent of dxFeed-specific events. Switching providers may require changes to authentication, subscriptions, event handling and entitlements; the effort will depend on your integration. Keep a provider adapter in your codebase and avoid making proprietary event names the only internal representation.

The homepage says dxFeed stores millisecond time-stamped tick data and processes up to 10Tb of raw data daily, but that does not establish service-level guarantees, client limits or end-to-end latency. The vendor does not publish rate limits or reliability figures in the homepage text. Ask for applicable SLAs, connection and subscription limits, recovery behavior, security controls, compliance terms and incident communications before making the feed a dependency.

Reddit discussions describe differences in DOM or order-flow displays between dxFeed and Rithmic; commenters suggest collection, consolidation, platform settings or update rates could explain discrepancies, but do not establish a cause. In another thread, users report that VIX subscriptions returned no data and that approval from OPRA was required; commenters also describe unclear API errors. Treat these as reports, not general guarantees. Validate entitlements and compare live and historical results for the exact instruments and platform you plan to support.