Repolode
Submit
elastic

elastic/elastic-labs

Executable Python notebooks, sample apps and resources for trying Elasticsearch across search, observability and security use cases.

View on GitHub
Stars
1,140
Forks
280
Open issues
52
Last commit
4 days ago

What is it?

elastic-labs is a collection of runnable Python notebooks and sample applications from Elastic's Search team, covering how to use Elasticsearch as a vector database, build RAG and chatbot apps, and run hybrid search. It also includes material on OpenTelemetry-based observability with ES|QL queries and on security detection engineering and LLM application attack testing.

The notebooks cover kNN and hybrid search with reciprocal rank fusion, ELSER and multilingual E5 semantic search, query rules, synonyms, learning to rank, and semantic reranking. Integration examples connect Elasticsearch with OpenAI, Hugging Face, Cohere, Amazon Bedrock, Anthropic and LangChain. Three example apps are included: a Chatbot RAG App, an Internal Knowledge Search app and a Relevance Workbench.

The project is Apache-2.0 licensed and maintained by Elastic's Search team. Sample app code is not covered by Elastic's support services.

Key features

  • 01Runs Elasticsearch as a vector database with kNN and hybrid search.
  • 02Provides notebooks for ELSER and multilingual E5 semantic search.
  • 03Includes RAG and chatbot example apps built on Elasticsearch.
  • 04Shows integrations with OpenAI, Hugging Face, Cohere, Amazon Bedrock, Anthropic and LangChain.
  • 05Covers OpenTelemetry observability and ES|QL log analysis examples.
  • 06Includes security detection queries using ES|QL and EQL.

Use cases

  • Learning how to build semantic search or RAG features on Elasticsearch.
  • Prototyping a chatbot that answers questions over internal documents.
  • Experimenting with query rules, synonyms and reranking to tune search relevance.
  • Exploring ES|QL queries for logs, observability data or threat hunting.

Quick start

  1. 1

    Open a notebook from the notebooks folder, such as the search quick start, in your browser via Colab or locally

  2. 2

    Run the notebook cells against your own Elasticsearch deployment

Things to know

  • This is a collection of examples and notebooks, not a single installable library.
  • The notebooks assume you have an Elasticsearch deployment to connect to.
  • Elastic support does not cover the sample application code in this repository.

In collections

OpenTelemetry & observability platforms

Topics

#elasticsearch#vector-search#rag#semantic-search#jupyter-notebook#langchain#observability#hybrid-search#ai#applications

FAQ

What is elastic-labs?
Executable Python notebooks, sample apps and resources for trying Elasticsearch across search, observability and security use cases. elastic-labs is a collection of runnable Python notebooks and sample applications from Elastic's Search team, covering how to use Elasticsearch as a vector database, build RAG and chatbot apps, and run hybrid search. It also includes material on OpenTelemetry-based observability with ES|QL queries and on security detection engineering and LLM application attack testing.
Is elastic-labs free to use? What license does it have?
elastic-labs is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is elastic-labs still maintained?
Actively maintained. The last commit to elastic-labs was on Nov 3, 2026.
What can I use elastic-labs for?
Learning how to build semantic search or RAG features on Elasticsearch. Prototyping a chatbot that answers questions over internal documents. Experimenting with query rules, synonyms and reranking to tune search relevance. Exploring ES|QL queries for logs, observability data or threat hunting.
How do I install elastic-labs?
Open a notebook from the notebooks folder, such as the search quick start, in your browser via Colab or locally Run the notebook cells against your own Elasticsearch deployment
Where can I get elastic-labs?
The source code and installation instructions are on GitHub: https://github.com/elastic/elastic-labs

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects