Progress launches MarkLogic Server 12 with new AI capabilities

Progress, an AI-powered infrastructure software provider, has announced the release of Progress MarkLogic Server 12, which introduces new generative AI (GenAI) capabilities. These enhancements aim to increase the accuracy of AI-generated responses and simplify the development of secure GenAI-enhanced applications.

The new features in MarkLogic Server 12 build on the existing search capabilities, adding native vector support and relevance-based algorithms. This aims to provide advanced capabilities for customers to harness AI within their enterprise environments. The latest release also allows organisations to integrate GenAI in their applications more easily through a flexible Retrieval Augmented Generation (RAG) pattern. This ensures that large language models (LLMs) are securely grounded in private data, enabling fact-based AI-generated query responses.

John Ainsworth, EVP and General Manager, Application and Data Platform, Progress, commented on the release, stating, "As AI becomes increasingly integrated into the enterprise, it is crucial to not only improve the precision of the results delivered but also to safeguard the enterprise's valuable data. With the latest release of MarkLogic Server, our customers have access to a dependable database management system capable of generating top-quality AI responses, while also protecting their reputation and intellectual property investment."

The new native vector support in MarkLogic Server 12 enables large-scale similarity searches for AI systems on unstructured data. Users can perform vector operations directly within the MarkLogic Server to improve GenAI retrieval accuracy. Additionally, the BM25 relevance ranking feature improves document ranking, enabling the most relevant search results to be returned at the top. This feature combines best-match and vector search to re-rank results for optimal search relevance. Organisations can merge and re-rank search result scores through a single expression via its unified Optic API, which parses the content most relevant to a user query to the LLM model.

Further enhancements include the graph shortest path feature, which improves contextual relevance of results and response time in AI-enhanced recommendation systems using semantic searches against knowledge graphs or semantic data in graph RAG architectures. This is beneficial for solving real-world problems related to route planning, network optimisation, and resource allocation.

Progress has also introduced a new application, MarkLogic Flux, which enhances data movement and transformation capabilities. This includes support for more data sources, such as relational databases and cloud storage systems like Amazon S3, and improves performance through faster ingestion of high-volume data. These features support various operational, analytical, and AI workloads.

The Progress MarkLogic platform, which includes the MarkLogic Server, aims to combine multi-model data management with real-time, relevance-based search and semantic capabilities. This provides an adaptable and secure foundation for GenAI solutions within enterprises. The flexible RAG architecture augments LLMs with domain-specific knowledge to increase response accuracy, democratising access to information across the organisation and powering AI-enhanced applications.