Google Cloud made expanded Knowledge Catalog semantic-model support generally available, integrating definitions from Looker, dbt, Snowflake, Databricks, and SAP.

Google Cloud announced general availability of expanded semantic-model support in Knowledge Catalog. Native ingestion covers Looker LookML, BigQuery Graph including BigQuery measures, dbt MetricFlow, Snowflake Semantic Views, Databricks Metric Views, and SAP Core Schema Notation.

Knowledge Catalog connects those definitions with schemas, data lineage, data profiles, business glossaries, and other metadata in a single context graph. Google says this works without moving the underlying data or rewriting business logic in a new language, and conflicting definitions retain their provenance so data stewards can reconcile them.

The context graph can ground Gemini agent, BigQuery Conversational Analytics, and custom agents in existing metrics, filters, and join paths. Google also describes Gemini-powered self-learning enrichment that extracts context from operational playbooks, meeting notes, chat threads, code, and other unstructured sources, turning missing or outdated information into governed, reusable context sets.

Based on Google’s internal benchmarks, the company reports nearly 4× greater accuracy, up to 33% lower latency, and up to 75% fewer tokens when grounding agents in Knowledge Catalog. These are vendor-reported benchmark results, not independently verified outcomes or guaranteed improvements across every workload; the supplied excerpt does not provide the full evaluation methodology.