RELEVANCE

Personalized search, unparalleled relevance

With powerful out-of-the box search relevance, Elastic® provides all the tools you need to build AI-powered search experiences that help users find exactly what they need. The Elasticsearch Relevance Engine™, state of the art machine learning, and relevance tuning tools help you analyze, optimize, and personalize even further.

Video thumbnail

Learn about the Elasticsearch Relevance Engine (ESRE) for creating AI search applications that integrate with LLMs and generative AI.

Read blog

See how easy it is to get started with setting up the Elasticsearch Relevance Engine.

See quick start video

Get an introduction to Elasticsearch's advanced relevance ranking toolbox.

Watch webinar

AI-POWERED RELEVANCE

Create AI search applications and integrate with large language models with the Elasticsearch Relevance Engine. Use industry leading advanced relevance ranking features like BM25f for hybrid search, native vector search, Elastic's proprietary ML model for semantic search across domains, and hybrid ranking using RRF to enter a new era of contextual relevance.

ELSER, INFERENCE API

Model selection made easy

Accelerate your RAG implementations with the Elastic Learned Sparse EncodeR (ELSER) as a reliable starting point. Additionally, Elastic's Inference API streamlines code and multi-cloud inference management. Whether you use ELSER or embeddings from OpenAI, Hugging Face, Cohere, or others for RAG workloads, one API call ensures clean code for managing hybrid inference deployment.

LEARNING TO RANK

The most relevant search engine for RAG

Rerankers apply machine learning models to fine tune your search results, and bring the most relevant results to the top based on user preferences and signals. Learning to Rank (LTR) is native to Elastic and supports RAG use cases — feeding the most relevant results to LLMs as context.

QUERY RULES AND SYNONYMS API

Optimize search performance

Provide customizable instructions through metadata for more control of search results in response to targeted queries. Query rules in Elasticsearch help you promote high-priority content to end-users for specific use cases. Additionally, you can simplify organizing and updating related words for website searches using the synonyms management API.

Fine-tune your search relevance model

Elasticsearch query language supports advanced search techniques (full-text, sparse/dense vector search), along with hybrid search using reciprocal rank fusion (RRF). Combine this with filtering, boosting and rescoring methods, and you're able to further fine-tune your search relevance model, customizing to your needs.

HYPER-RELEVANCE

Harness the power of machine learning

Whether you're adding new concepts to broaden the impact of your search or seeking new ways to improve search accuracy, machine learning can augment search and business insights to enhance your search applications and customer experience. Improve semantic relevance with generative AI, vector search, support for NLP transformer models, and third-party model management.