June 2026 Summaries

Weaviate v1.38 introduces several new features and updates, including the general availability of the HFresh disk-based vector index and the built-in Model Context Protocol (MCP) server. The HFresh index, inspired by the SPFresh algorithm, efficiently manages memory and latency for streaming workloads by storing vectors on disk and maintaining a small in-memory index. The MCP server enables seamless integration with LLMs and AI agents for hybrid searches and object management. The latest version also includes a re-architected async replication process for improved consistency across clusters, running from a single scheduler and shared worker pool by default. Additionally, two preview features have been introduced: the Boost API for query-time rescoring and Nested Object Filtering, which allows filtering on nested object properties. The release also boasts multiple performance improvements and community contributions, making it a significant update for users of the platform.

Many vector database pilots encounter issues at the data ingestion stage rather than during search, with challenges often arising when scaling from small datasets to much larger ones. This guide provides insights into overcoming common problems such as rate limits, partial batch failures, and memory issues, emphasizing the importance of server-side batching, error handling, and data type decisions when using Weaviate for data import. It discusses the use of server-side batching that dynamically adjusts based on server workload, and highlights strategies for error handling and retries to ensure successful data import. The guide also covers the benefits of using the blobHash data type for media to reduce storage requirements and enable efficient re-vectorization checks. Additionally, it explores options for multimodal data ingestion, particularly for image-based document retrieval without needing an OCR pipeline, and offers practical advice on making schema decisions to avoid costly post-import fixes. A Weaviate Cloud trial with Weaviate Embeddings is recommended for users to experiment with these features in a low-friction environment.

Weaviate has historically been open source and free, allowing users to run it on their local machines or clouds without cost, but recent developments have shifted Weaviate from being solely a database to a comprehensive platform. This evolution includes the introduction of the Query Agent, which enables natural language queries, and Engram, a managed memory layer that enhances user interaction over time. Recognizing the unique nature of these services, Weaviate has introduced free tiers for the entire suite of products on Weaviate Cloud, including the previously time-limited managed database, ensuring users can access and utilize these tools without initial costs or time constraints. This change allows users to build and develop their projects freely until they require scaling, at which point they can transition seamlessly to a pay-as-you-go model. This new approach emphasizes providing a lasting, unrestricted free experience rather than a temporary one, encouraging users to explore and innovate with Weaviate's offerings.

Engram, now generally available, is a managed memory and context service designed to enhance the functionality of agents by turning raw, noisy events into structured memories, facilitating efficient workflow orchestration and decision-making. It addresses challenges in agent memory management such as long-context degradation, messy raw data, and multi-agent context fragmentation by providing a systematic approach to memory as an infrastructure component, rather than a superficial addition. Engram utilizes Weaviate's hybrid semantic and keyword retrieval, with asynchronous pipelines that extract, deduplicate, and reconcile information to maintain a clean memory state. The service supports personalization, continual learning, and multi-agent state through ready-to-deploy templates, while allowing teams to customize their pipelines without leaving the platform. Engram is particularly beneficial for teams with agents that require persistent memory across sessions, need to improve from feedback, or involve multi-agent systems needing shared state, and it is available in Weaviate Cloud with a free tier to help users begin enhancing their agents' capabilities.