Skip to main content
GDELT Cloud plans are packaged around use-cases. Every plan includes the core data — stories, coded events (CAMEO+/ACLED), resolved entities, and Global Energy Monitor assets — and higher plans add data modules on top. Current prices and limits always live on the pricing page; this page documents how module access maps to the API.

​
Data modules

Risk Context is reference context, not a compliance product. Matches are cited, best-effort signals — not a compliance determination, not a system of record, and not legal advice. See the methodology for provenance and appropriate-use boundaries.

​
Which plans include which capabilities

Plans are a strict ladder — each tier includes everything below it — and they are grouped the same way the API Arena groups endpoints: what we produce, then what we ingest and fuse, then what we derive.
  • Core Data — coded events, clustered stories, resolved entities, energy & heavy-industry assets, and the fused facilities directory. Included on every tier, including Free.
  • Core Analytics — cross-language media tone and Share of Voice.
  • Open Feeds — government exposure (USAspending + FARA), SEC filings, AI & compute (Epoch AI), markets & macro (FRED/ALFRED), risk & screening (OFAC/BIS, AidData), and maritime AIS.
  • Intel — Monitoring Briefs, alerts, the research agent, and Atlas indices.
  • Enterprise additionally unlocks point-in-time (as_of) data, custom QU and request-rate limits, and per-contract seat counts.
Sizing QU. Our data refreshes hourly, so keeping one entity, topic or query current costs about 750 QU/month — polling faster returns the same answer. Divide a plan’s QU by 750 for the number of things you can keep current (~3 on Builder, ~27 on Analyst, ~137 on Intelligence). Briefs are metered separately and never draw down query units.
Retired tiers. Monitor, Media Intelligence, Corporate & Supply Chain and Global Intelligence are no longer sold. Existing subscribers keep their plan, price and every feature — these tiers continue to resolve exactly as before, and the à-la-carte module storefront has been retired in favour of the bundled ladder above.

​
How gating behaves in the API

  • Every /api/v2 route authenticates with your API key and evaluates your plan’s module flags. Calling a module endpoint without the module returns 403 with code: "PLAN_REQUIRED" and a message naming the module. Your key is not blocked — non-module endpoints keep working.
  • MCP inherits the same gates. MCP tools call the same /api/v2 routes with your API key, so tool availability follows your plan automatically.
  • Entity pages and /api/v2/entities/{id} return a per-source sources map: sources your plan includes report real availability; others return a uniform locked state.
  • Share of Voice requests are bounded for reliability: at most 25 entities per request and a 92-day window (ENTITY_LIMIT_EXCEEDED / WINDOW_TOO_LARGE otherwise).

​
Rate limits and metering

Query Units (QU) meter usage: 1 QU per REST/MCP call, 5 QU per in-app Agent tool call. Each plan carries a monthly QU allowance, per-minute rate limit (RPM), alert and Brief allowances — see pricing for current numbers. Briefs meter separately in Brief Units and never draw down QU.