GDELT Cloud uses Tavily to fetch and parse article bodies across its ingest pipeline for event coding. Tavily processes article URLs fetched during ingest. Parallel Web Systems provides fallback extraction of public article pages when Tavily cannot retrieve them.
The disclosure also identifies prior research-agent use through retained search queries and target URLs. Hosted-agent web research is temporarily disabled.
The V2 entity-tone scorer supports high-volume concurrent processing with a per-run Tavily fallback budget. This lets hourly dogfood ingest cover thousands of queued entity-story cases without making web extraction the critical path.
GDELT Cloud identifies Tavily as its article extraction subprocessor for ingest and event coding. Hosted-agent web research is temporarily disabled. It retains search queries and target URLs from prior research-agent runs. Parallel Web Systems provides fallback extraction when Tavily cannot retrieve an article.
Subprocessors & Data Processing Vendors15 Oct 2026 · GDELT Cloud · Web pageGDELT Cloud identifies Tavily as its article extraction subprocessor for ingest and event coding. Hosted-agent web research is temporarily disabled. It retains search queries and target URLs from prior research-agent runs. Parallel Web Systems provides fallback extraction when Tavily cannot retrieve an article.
Changelog | GDELT Cloud
19 Oct 2026 · gdeltcloud.com · Web page
GDELT Cloud identifies Tavily as its provider for web search and content extraction. Its V2 entity-tone scorer supports high-volume concurrent processing with a per-run Tavily fallback budget, allowing hourly dogfood ingest to cover thousands of queued entity-story cases without making web extraction the critical path.
Changelog | GDELT Cloud19 Oct 2026 · gdeltcloud.com · Web pageGDELT Cloud identifies Tavily as its provider for web search and content extraction. Its V2 entity-tone scorer supports high-volume concurrent processing with a per-run Tavily fallback budget, allowing hourly dogfood ingest to cover thousands of queued entity-story cases without making web extraction the critical path.
GDELT Cloud turns global news and public records into structured event intelligence for developers, researchers and teams monitoring risk and operations. Its products include a REST API, an MCP server for AI agents, and scheduled Monitors that deliver updates by email or signed webhook. The platform classifies events, groups related articles into Stories, links people and organizations, and tracks developments through Situations. It also connects reporting with company filings, economic indicators and physical infrastructure data for applications such as supply-chain monitoring, news products and country research.
Provides API and agent-native access to structured global event intelligence. · Infrastructure
How they use Tavily
GDELT Cloud uses Tavily to fetch and parse article bodies across its ingest pipeline for event coding. Tavily processes article URLs fetched during ingest. Parallel Web Systems provides fallback extraction of public article pages when Tavily cannot retrieve them.
The disclosure also identifies prior research-agent use through retained search queries and target URLs. Hosted-agent web research is temporarily disabled.
The V2 entity-tone scorer supports high-volume concurrent processing with a per-run Tavily fallback budget. This lets hourly dogfood ingest cover thousands of queued entity-story cases without making web extraction the critical path.
GDELT Cloud identifies Tavily as its article extraction subprocessor for ingest and event coding. Hosted-agent web research is temporarily disabled. It retains search queries and target URLs from prior research-agent runs. Parallel Web Systems provides fallback extraction when Tavily cannot retrieve an article.
Subprocessors & Data Processing Vendors15 Oct 2026 · GDELT Cloud · Web pageGDELT Cloud identifies Tavily as its article extraction subprocessor for ingest and event coding. Hosted-agent web research is temporarily disabled. It retains search queries and target URLs from prior research-agent runs. Parallel Web Systems provides fallback extraction when Tavily cannot retrieve an article.
Changelog | GDELT Cloud
19 Oct 2026 · gdeltcloud.com · Web page
GDELT Cloud identifies Tavily as its provider for web search and content extraction. Its V2 entity-tone scorer supports high-volume concurrent processing with a per-run Tavily fallback budget, allowing hourly dogfood ingest to cover thousands of queued entity-story cases without making web extraction the critical path.
Changelog | GDELT Cloud19 Oct 2026 · gdeltcloud.com · Web pageGDELT Cloud identifies Tavily as its provider for web search and content extraction. Its V2 entity-tone scorer supports high-volume concurrent processing with a per-run Tavily fallback budget, allowing hourly dogfood ingest to cover thousands of queued entity-story cases without making web extraction the critical path.
GDELT Cloud turns global news and public records into structured event intelligence for developers, researchers and teams monitoring risk and operations. Its products include a REST API, an MCP server for AI agents, and scheduled Monitors that deliver updates by email or signed webhook. The platform classifies events, groups related articles into Stories, links people and organizations, and tracks developments through Situations. It also connects reporting with company filings, economic indicators and physical infrastructure data for applications such as supply-chain monitoring, news products and country research.
