﻿---
title: Generative AI connectors
description: Use these connectors to connect to third-party large language model (LLM) services and Elastic's own LLM offerings. 
url: https://www.elastic.co/elastic/docs-builder/docs/3716/reference/kibana/connectors-kibana/gen-ai-connectors
products:
  - Kibana
applies_to:
  - Elastic Cloud Serverless: Deprecated
  - Elastic Stack: Planned for deprecation
---

# Generative AI connectors
Use these connectors to connect to third-party large language model (LLM) services and Elastic's own LLM offerings.

## Available connectors

<important applies-to="Elastic Cloud Serverless: Deprecated, Elastic Stack: Planned for deprecation">
  This connector is deprecated and is being progressively removed from the create connector UI. Existing connectors and their rule actions continue to work.For new AI integrations, use Elasticsearch inference endpoints. Migrate existing LLM connectors and related rule actions before the future removal.
</important>

- [AI Connector](https://www.elastic.co/elastic/docs-builder/docs/3716/reference/kibana/connectors-kibana/ai-connector): Connect to third-party LLM services including Amazon Bedrock, Azure, Google Gemini, OpenAI, and Elastic Inference Service.
- [Amazon Bedrock](https://www.elastic.co/elastic/docs-builder/docs/3716/reference/kibana/connectors-kibana/bedrock-action-type): Send a request to Amazon Bedrock.
- [Elastic Managed LLMs](https://www.elastic.co/elastic/docs-builder/docs/3716/reference/kibana/connectors-kibana/elastic-managed-llm): Send a request to Elastic Managed LLMs.
- [Google Gemini](https://www.elastic.co/elastic/docs-builder/docs/3716/reference/kibana/connectors-kibana/gemini-action-type): Send a request to Google Gemini.
- [OpenAI](https://www.elastic.co/elastic/docs-builder/docs/3716/reference/kibana/connectors-kibana/openai-action-type): Send a request to OpenAI.

**External MCP Server**
- [MCP](https://www.elastic.co/elastic/docs-builder/docs/3716/reference/kibana/connectors-kibana/mcp-action-type): Connect to MCP servers and call their tools. <applies-to>Elastic Stack: Preview in 9.3</applies-to> <applies-to>Elastic Cloud Serverless: Preview</applies-to>

<important>
  Connecting to LLM providers through a proxy is in technical preview. If you use a proxy, it should support streaming and be SSE-compatible. Elastic only parses streamed responses.To check if problems are caused by using a proxy, you can test your LLM service without using a proxy.
</important>