Spanner Omni, now GA: A distributed, multi-model database that you can deploy anywhere

Spanner Omni, the deploy-anywhere version of Google’s Spanner database, is now generally available after its debut at Google Cloud Next ’26. The service extends Spanner’s distributed SQL capabilities beyond Google Cloud, allowing enterprises to run demanding production workloads in private data centers or other cloud environments. It combines horizontal scalability with ACID compliance and strong consistency, maintaining the core features of the fully managed Spanner service while offering deployment flexibility.

The database supports multiple data models—including SQL, graph, key-value, full-text search, vector search, and analytical processing—within a single system. Spanner Omni is designed for agentic AI applications, enabling teams to build once and deploy across Google Cloud, on-premises, or other clouds. It introduces dedicated worker nodes to offload resource-intensive operations, enhancing performance for core database workloads in self-managed environments.

Attio, an AI-native CRM company based in London, has adopted Spanner Omni for its agentic application environment. The company migrated to a containerized Spanner Omni deployment alongside managed Spanner, citing accelerated production velocity and the ability to handle complex, mission-critical workflows such as Spanner queues. Attio’s CTO highlighted the enhanced scale and confidence enabled by Spanner Omni’s capabilities across all environments.

Spanner Omni offers two licensing tiers for production deployments and includes robust security features like TLS encryption, authentication, authorization, and audit logging. It supports advanced AI workloads with native vector search, graph capabilities, and integration via the Model Context Protocol (MCP) Toolbox for agentic systems. While some Google Cloud-specific integrations are excluded, Google is actively closing feature gaps between Spanner Omni and the fully managed service.