Google Cloud announced the general availability of BigQuery Graph, bringing native graph capabilities directly into the BigQuery data warehouse. The key problem addressed is the traditional need to extract data into standalone graph databases, creating silos and operational overhead. BigQuery Graph eliminates ETL by using ISO-standard Graph Query Language (GQL) alongside SQL, with traversals running natively on petabyte-scale data while inheriting BigQuery’s row- and column-level security and AI functions such as BigQuery ML.

During preview, data teams used BigQuery Graph for threat and fraud detection, supply chain digital twins, identity resolution, knowledge graphs for AI agent grounding, and network lineage management. The GA release focuses on a faster and broader graph engine: path-finding optimizations make GQL 2x faster since preview and undirected traversal 100x faster, with improved cycle detection in ACYCLIC and TRAIL modes. The new borderless lakehouse capability allows a single BigQuery Graph to span native BigQuery tables and open Iceberg tables in other clouds (Databricks Unity Catalog, AWS Glue, Snowflake) without copying data or building ETL pipelines. More expressive GQL is supported via the CALL statement and extended subquery support, enabling reusable named functions that agents can call as tools.

Several agentic ecosystem features are introduced or rolling out. Conversational analytics lets users explore graphs in natural language by translating questions into SQL or GQL, with visualizations for path-based answers. A new agent skill packages BigQuery Graph expertise (schema design, GQL pattern matching) into tools like Antigravity, VS Code, Claude Code, and Codex via the Google Cloud Data Agent Kit extension; the skill can soon propose and verify node/edge models against real data with match rates. For agent auditing, context graph in BigQuery Agent Analytics captures each agent action as a typed, queryable context graph stored in BigQuery Graph, enabling explainable traces of decisions. Some features remain in preview or are rolling out over the coming weeks. The announcement includes quotes from Thales (Pete Rubio), Yahoo (Mikul Bhatt), and Workerbee (Heiko Roth) as early adopters.