Weaviate Advances Vector Database Technology to Power Fast-Growing AI Engineering Stacks

Weaviate Enterprise Edition and Weaviate Platform Edition introduce new operational capabilities and support for AI Engineering teams running AI data layers that form the foundation for AI initiatives across the organization.

Key Takeaways

  • Weaviate announces general availability of the Enterprise Edition and the Platform Edition with Weaviate 1.40, adding capabilities that simplify security, scaling, and high availability for enterprise AI Engineering teams.
  • The new editions introduce operational features including Shard Self-Recovery, Deduplicated Backups, and Namespaces to automate recovery, reduce backup storage and time, and provide control-plane and data isolation for shared clusters.
  • Weaviate deprecates GraphQL in favor of native client libraries and gRPC and REST APIs, with new Cloud clusters already created without GraphQL and shutdowns scheduled for December 1, 2026 for clusters that have not used it in the past 30 days and May 1, 2027 for all remaining clusters, followed by default disablement in the first self-hosted minor release after May 1, 2027.

Weaviate (https://weaviate.io/), the AI-native vector database company, today announced the general availability of the Weaviate Enterprise Edition and the Weaviate Platform Edition. Introduced with Weaviate version 1.40, the new offerings add new security, scaling, and high availability capabilities and services that make it easier for AI Engineering teams to centrally support the rapidly growing number of AI systems running across the enterprise.

Weaviate is an open-source, AI-native vector database purpose-built for AI Engineering teams. It stores vector embeddings together with the data objects they represent, and serves semantic, keyword and hybrid search for generative AI, retrieval-augmented generation (RAG), agentic AI, and inference workloads. With over 1 billion open-source downloads, Weaviate runs on-premises, in any cloud, or as a fully managed service on Weaviate Cloud.

Vector databases are the heart of the AI Engineering stack

Enterprises are consolidating siloed AI projects into centralized AI Engineering teams. The rise of AI Engineering within the enterprise minimizes AI silos and enables greater AI agility and trust by centralizing operational expertise and technology standards. The AI Engineering tech stack serves many AI initiatives at once, 24 hours a day. At the heart of that stack, a native vector database supplies the context that GenAI, RAG and agentic applications depend on.

"AI Engineering teams have stopped treating the vector database as a backend detail. It is now core infrastructure, chosen with the same rigor as a primary datastore," said Bob van Luijt, co-founder and CEO of Weaviate. "The Enterprise and Platform Editions give those teams the developer experience they already love in open-source Weaviate, plus the performance, scale and operational resilience to run it in mission-critical production around the clock."

Advanced features for secure, fast, highly available operations

The new Weaviate database editions include features that simplify operations for AI Engineering teams. Weaviate Enterprise Edition introduces performance, security, and operational features for teams operating AI applications at scale. Weaviate Platform is designed for teams operating Weaviate as a shared service in support of many AI projects. New features include:

  • Shard Self-Recovery — Automatically restores a shard whose data is missing on a node, such as after a disk replacement or a lost volume, by copying it from a healthy replica. Once enabled, recovery starts automatically when the node restarts, with no manual steps. It builds on the replication that's included free in the Community Edition.

  • Deduplicated Backups — Archives each converged shard once instead of once per replica, then restores it to every replica, cutting backup storage and time for replicated clusters.

  • Namespaces — The Weaviate Platform Edition adds control-plane and data isolation between users on a shared cluster, so one platform team can safely serve many internal teams and applications from the same deployment.

For a detailed comparison of Weaviate database edition functionality, visit https://weaviate.io/pricing#self-managed.

No change for open-source and Weaviate Cloud users. The open-source Weaviate database, now called the Community Edition, remains licensed under BSD-3-Clause, with no features removed and no limits added. All editions ship in one binary; a license key enables Enterprise or Platform Edition features at runtime for self-hosted deployments. Weaviate Cloud users need no license key. Self-hosted teams can request a free 30-day trial key.

Why AI Engineering teams choose Weaviate

Unlike hybrid databases such as pgvector (PostgreSQL) and OpenSearch, which add vector search to an engine designed for other workloads, Weaviate is purpose-built for vector data management. And unlike other vector databases, Weaviate’s advanced feature set and performance make it capable of supporting the widest range of AI use cases in the enterprise.

  • Price-performance tuned to every use case. Weaviate offers a broad set of search, indexing and compression options so teams can match cost and latency to each workload, whether GenAI, RAG, agentic AI, or inference. Search options include vector, keyword (BM25), hybrid, filtered, and multi-vector search. Multiple vector index types and vector quantization methods, including state of the art rotational quantization (RQ), let teams trade memory for recall per collection.

  • Vectors and data objects in one database. Weaviate stores vector embeddings together with the data objects to which they relate. Teams do not need a separate relational database or document store such as PostgreSQL or MongoDB to hold object data, which removes a synchronization layer and simplifies the stack.

  • Flexible deployment, on-premises or in the cloud, with support for SOC II, ISO 27001 and HIPAA environments. Weaviate runs self-hosted on-premises or in any public cloud using the same open-source image, or as a fully managed service on Weaviate Cloud, so AI Engineering teams can meet data-residency, security and cost requirements for each application.

  • A proven open-source foundation. With over 1 billion downloads, open-source Weaviate has one of the largest user communities of any vector database, and native client libraries for Python, TypeScript/JavaScript, Java, Go and C# keep developers productive.

Hello Weaviate Enterprise, Goodbye GraphQL

Alongside the Enterprise and Platform Edition functionality, Weaviate is ending GraphQL support in the Weaviate database. The GraphQL API has been deprecated in favor of Weaviate's native client libraries and gRPC and REST APIs, which offer faster queries and a simpler developer experience. New Weaviate Cloud clusters are already created without GraphQL. Weaviate Cloud switches GraphQL off on December 1, 2026 for clusters that have not used it in the past 30 days, and on May 1, 2027 for all remaining clusters. For self-hosted deployments, the first minor release after May 1, 2027 disables GraphQL by default, and a later release removes it. A migration guide explains how to find and replace GraphQL usage.

Availability

The Weaviate Enterprise Edition and Platform Edition are available now with Weaviate 1.40. Visit weaviate.io to create a Weaviate Cloud account or to get started with Weaviate open source.


About Weaviate

Weaviate is an open-source, AI-native vector database that helps AI Engineering teams build and run GenAI, RAG, agentic and search applications at scale. Weaviate stores vectors and data objects together, offers vector, keyword, and hybrid search with built-in compression and multi-tenancy, and runs on-premises, in any cloud, or fully managed on Weaviate Cloud. Open-source Weaviate has been downloaded over 1 billion times. Learn more at weaviate.io.

SOURCE: Weaviate

Key Takeaways

  • Weaviate announces general availability of the Enterprise Edition and the Platform Edition with Weaviate 1.40, adding capabilities that simplify security, scaling, and high availability for enterprise AI Engineering teams.
  • The new editions introduce operational features including Shard Self-Recovery, Deduplicated Backups, and Namespaces to automate recovery, reduce backup storage and time, and provide control-plane and data isolation for shared clusters.
  • Weaviate deprecates GraphQL in favor of native client libraries and gRPC and REST APIs, with new Cloud clusters already created without GraphQL and shutdowns scheduled for December 1, 2026 for clusters that have not used it in the past 30 days and May 1, 2027 for all remaining clusters, followed by default disablement in the first self-hosted minor release after May 1, 2027.

Media Gallery

Quotes

“
"AI Engineering teams have stopped treating the vector database as a backend detail. It is now core infrastructure, chosen with the same rigor as a...
Bob van Luijt — Co-founder and CEO, Weaviate

Contacts

Andy Ellicott
andy@weaviate.io
PR

FAQ

The Weaviate Enterprise Edition enhances security, scaling, and operational resilience, while the Platform Edition provides shared service functionality for multiple AI projects, featuring shard self-recovery, deduplicated backups, and namespaces for data isolation.

Teams can access these editions by downloading Weaviate 1.40, which includes both versions in a single binary. A license key activates specific features for self-hosted deployments, while Weaviate Cloud includes Enterprise features automatically.

Weaviate is purpose-built for vector data management, avoiding the complexity of hybrid databases by storing vector embeddings alongside their data objects and simplifying infrastructure without needing separate relational or document stores.

Yes, the open-source Weaviate, now referred to as the Community Edition, remains unchanged with no features removed or limits added.

Weaviate is phasing out GraphQL support, with its APIs transitioning to gRPC and REST, aiming for faster queries and improved developer experience, and will disable GraphQL in future self-hosted releases after May 1, 2027.

Weaviate is purpose-built for vector data management and supports many AI use cases including GenAI, RAG, agentic AI and inference, from one platform. Advanced features, flexible deployment, multi-tenancy and large open-source community let a centralized AI Engineering team run customer-facing AI applications around the clock.