From test to record.
OONI publishes network measurements. Voidly connects selected observations to source-linked records.
OONI is an upstream source for Voidly. Use OONI directly for its measurement records and test methodology. A Voidly record may instead be based on another source.
Same question. Both perspectives.
Capabilities in context, with source links. No color-coded winners or unmatched accuracy scores.
Documentation reviewed . The dates and scope of the underlying evidence remain separate. Unknown coverage or unverified interface absence is not scored as “no.”
OONI
OONI is a nonprofit free-software project founded in 2012. It publishes Probe tools, measurement access through Explorer/API and research reports. Its test documentation explains the distinct web, messaging, circumvention and performance checks.
Explore OONI ↗Voidly
Voidly adds an incident/index layer across OONI, CensoredPlanet, IODA and its own probes; Citizen Lab lists supply target/category context rather than independent measurements. APIs, feeds, alert options and citations expose those records. Follow each record’s actual source instead of assuming all contain OONI evidence.
How we measure ↗API docs ↗Follow the work
Choose by the task, then check the actual record. Each source has a different role.
Find relevant records, then inspect the actual measurement source and observation window. Follow OONI links where present; IR-2026-0142 is an IODA example, not an OONI-backed block example.
Use the documented accessibility API/MCP for a source-backed signal. Inspect network evidence, provider restrictions and freshness separately; it cannot guarantee an individual user’s access.
Review OONI Probe’s web-connectivity method and data policy before choosing to run a measurement on a network you control. This page does not initiate tests or promise a Voidly on-demand probe.
Read OONI’s participation and data-publication guidance before volunteering. Voidly also documents probe participation; no unverified fleet-size comparison is needed to choose.
Use an incident’s readable ID and stable page alongside the underlying evidence. Credit whichever sources that record actually cites, and distinguish outage signals from a verified blocking mechanism.
Use the overlap
Start with the record or the measurement, depending on your question. When a Voidly record cites OONI, follow that link into Explorer and examine the test, network and time. Cite OONI for its data and identify Voidly’s additional interpretation separately.
The source trail
This page’s conclusions are editorial comparisons of the linked documentation, not a controlled comparative evaluation.
Voidly-original material is CC BY 4.0. Upstream data keeps its source license. Public access and licensing are different questions; current account-key allowances and endpoint limits are documented in Pricing. Current task/settlement release details are on Voidpay.
Historical text captured from the original source on 2026-09-06. That is a capture date, not an observation date. These earlier claims are preserved for review, may be stale or incorrect, and should not be copied as current facts. The corrected comparison above takes precedence.
How Voidly compares to OONI (Open Observatory of Network Interference). OONI runs the gold-standard probe network. Voidly aggregates OONI plus other sources into ML-classified incidents.
OONI runs the probes and publishes the raw data. Voidly aggregates, classifies, and indexes it. We are downstream of OONI, not a replacement.
Voidly is built on top of OONI data — they are an upstream source, not a competitor. We credit OONI in every dataset, page, and incident citation. For raw probe measurements and methodology, go to ooni.org directly.
| Feature | Voidly | OONI |
|---|---|---|
| Probe network | Partial100+ operators (volunteer + Vultr core) | Yes240+ countries, mass volunteer probes |
| Raw measurements | NoAggregated only — not a probe network | Yes3.5B+ raw measurements |
| Methodology rigor | PartialInherits OONI + adds ML classification | YesPeer-reviewed, gold standard |
| Data sources | YesOONI + CensoredPlanet + IODA + Citizen Lab | PartialOONI Probe network only |
| Citable incident IDs | YesIR-2026-0142 format, 7,500+ incidents | NoPer-measurement only |
| ML classification | YesGradientBoosting v3.3, LOCO mean F1 0.71 (0.63 on the countries with n≥30) | NoHeuristic anomaly detection |
| AI/MCP integration | Yes@voidly/mcp-server, 84 tools | NoNo MCP server |
| REST API | Yesapi.voidly.ai with auth + free tier | Yesapi.ooni.io aggregation API |
| Open source | PartialSDK + MCP open, backend mixed | YesFully open source |
| Non-profit | NoFor-profit, free tier + paid API | YesNon-profit, no commercial product |
| Years of data | Partial10-year OONI archive (1.6M records) | YesSince 2012, 14+ years |
| Real-time alerts | YesWebhooks, RSS, MCP, Sentinel forecasts | PartialOONI Run, no webhook subscriptions |
OONI (Open Observatory of Network Interference) is the foundational organization for empirical internet censorship measurement. They have run a global probe network since 2012 with over 3.5 billion measurements collected from volunteers in 240+ countries. Their methodology is rigorous, peer-reviewed, fully open-source (Probe app, Explorer, all backend tools), and their MAT (Measurement Aggregation Toolkit) is the standard reference for academic researchers. The OONI Explorer is the canonical place to view raw test results — anyone can drill down to a specific measurement and see the exact DNS resolution, TCP handshake, and HTTP response captured at the moment of testing. OONI is a non-profit and their data is the bedrock that every serious censorship researcher (including Voidly) relies on.
Voidly is an aggregation, classification, and indexing layer downstream of OONI. We ingest OONI's aggregation API every 6 hours alongside CensoredPlanet (Satellite + Hyperquack), IODA, Citizen Lab category mapping, and our own probe network. Where OONI publishes measurements, we publish incidents — clusters of measurements grouped into citable events with hash IDs (e.g. IR-2026-0142), ML-derived confidence scores, and structured evidence permalinks. We expose this as REST APIs, an MCP server, RSS feeds, BibTeX citations, and webhook subscriptions. We are optimized for AI agents and downstream applications; OONI is optimized for measurement integrity and researcher access. Different layers of the stack.
Start with Voidly to identify the incident (IR-2026-0142), then drill into OONI Explorer for the underlying raw measurements. Cite OONI for the data, Voidly for the classification.
Use Voidly MCP server. We pre-aggregate OONI + 3 other sources and expose them as 84 MCP tools. OONI does not have an MCP server.
Use OONI Probe (the app). Voidly does not run probes on demand — we consume aggregated probe data from OONI and others.
Install OONI Probe. Their volunteer network is far larger than Voidly's probe nodes. (Voidly accepts probe contributions but OONI is the right entry point.)
Use Voidly incident hash IDs (IR-2026-0142). Each links to evidence permalinks that cite the upstream OONI/CensoredPlanet/IODA measurements. Both are credited.
Voidly users should also use OONI. If you find an interesting incident in Voidly, the next step is usually to drill into OONI Explorer for the raw measurement data. Every Voidly incident links back to the underlying OONI measurements where applicable. We push researchers toward OONI for serious investigations — they are the source.
The old page’s final CTA described all data/ML/forecast reads as free without a key, applied one CC BY license, and described a handful of metered Voidpay endpoints. Current access, source licensing and Pay releases must be checked separately; that text was not a universal capability or pricing guarantee.
Complete historical comparison ↗Public data access
Start with the public dataset. Check its source terms, dates and endpoint limits.
Keep exploring
Country research and source-linked incident records. Two different ways to understand internet freedom.
Traffic trends and source-linked interference evidence. Align the time and network before comparing them.
