This repository provides executable Python notebooks and sample applications for leveraging Elasticsearch in modern search and AI/ML-powered applications. It targets developers and researchers looking to implement advanced search functionalities like vector databases, hybrid search, retrieval-augmented generation (RAG), and semantic search, integrating with popular AI frameworks.

How It Works

The project demonstrates Elasticsearch's capabilities as a vector database for storing embeddings and powering semantic search. It highlights advanced features such as the Elastic Learned Sparse Encoder (ELSER) and reciprocal rank fusion (RRF) for out-of-the-box, high-performance search without custom training. The examples showcase integration with LLM ecosystems like OpenAI, Hugging Face, and LangChain, positioning Elasticsearch as a robust backend for AI-driven applications.

Quick Start & Requirements

• Install/Run: Primarily through Python notebooks (e.g., Google Colab).
• Prerequisites: Python, Elasticsearch instance (local or cloud), potentially API keys for services like OpenAI or Hugging Face.
• Resources: Notebooks are designed for interactive use, with resource requirements dependent on the specific notebook's data and model usage.
• Links: Elasticsearch Labs, Contributing Guidelines

Highlighted Details

• Demonstrates Retrieval Augmented Generation (RAG) and question-answering (QA) use cases.
• Features hybrid search, semantic reranking, and keyword querying with filtering.
• Includes examples for document chunking and calculating tokens for semantic search.
• Showcases integrations with OpenAI, Anthropic Claude 3, Hugging Face, and LangChain.

Maintenance & Community

• Maintained by the Search team at Elastic.
• Support available via Elastic Discuss Forums (tag #esre-elasticsearch-relevance-engine) and the Elastic Community Slack (#search-esre-relevance-engine channel).

Licensing & Compatibility

• Licensed under the Apache License, version 2.0 (ALv2).
• Permissive license suitable for commercial use and integration with closed-source applications.

Limitations & Caveats

The repository contains sample application code, and official Elastic support services do not extend to this code. Some notebooks may require specific versions of dependencies or access to external AI services.