---
type: api
title: Inference
description: Inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Azure, Google AI Studio or Hugging Face. For built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models. However, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.
url: https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/group/endpoint-inference
resource: https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/group/endpoint-inference
products:
  - Elasticsearch
---

# Inference

`inference`

Inference APIs enable you to use certain services, such as built-in machine learning models (ELSER, E5), models uploaded through Eland, Cohere, OpenAI, Azure, Google AI Studio or Hugging Face. For built-in models and models uploaded through Eland, the inference APIs offer an alternative way to use and manage trained models. However, if you do not plan to use the inference APIs to use these models or if you want to use non-NLP models, use the machine learning trained model APIs.

- [Perform chat completion inference
](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-chat-completion-unified): `POST` `/_inference/chat_completion/{inference_id}/_stream`
- [Perform completion inference on the service](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-completion): `POST` `/_inference/completion/{inference_id}`
- [Get an inference endpoint
](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-get): `GET` `/_inference/{task_type}/{inference_id}`, `GET` `/_inference/{task_type}/_all`, `GET` `/_inference/{inference_id}`, `GET` `/_inference`
- [Create an inference endpoint
](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put): `PUT` `/_inference/{task_type}/{inference_id}`, `PUT` `/_inference/{inference_id}`
- [Perform inference on the service
](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-inference): `POST` `/_inference/{task_type}/{inference_id}`, `POST` `/_inference/{inference_id}`
- [Delete an inference endpoint
](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-delete): `DELETE` `/_inference/{task_type}/{inference_id}`, `DELETE` `/_inference/{inference_id}`
- [Create an AlibabaCloud AI Search inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-alibabacloud): `PUT` `/_inference/{task_type}/{alibabacloud_inference_id}`
- [Create an Amazon Bedrock inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-amazonbedrock): `PUT` `/_inference/{task_type}/{amazonbedrock_inference_id}`
- [Create an Amazon SageMaker inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-amazonsagemaker): `PUT` `/_inference/{task_type}/{amazonsagemaker_inference_id}`
- [Create an Anthropic inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-anthropic): `PUT` `/_inference/{task_type}/{anthropic_inference_id}`
- [Create an Azure AI studio inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-azureaistudio): `PUT` `/_inference/{task_type}/{azureaistudio_inference_id}`
- [Create an Azure OpenAI inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-azureopenai): `PUT` `/_inference/{task_type}/{azureopenai_inference_id}`
- [Create a Cohere inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-cohere): `PUT` `/_inference/{task_type}/{cohere_inference_id}`
- [Create a custom inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-custom): `PUT` `/_inference/{task_type}/{custom_inference_id}`
- [Create a DeepSeek inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-deepseek): `PUT` `/_inference/{task_type}/{deepseek_inference_id}`
- [Create an Elasticsearch inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-elasticsearch): `PUT` `/_inference/{task_type}/{elasticsearch_inference_id}`
- [Create an ELSER inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-elser): `PUT` `/_inference/{task_type}/{elser_inference_id}`
- [Create an Google AI Studio inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-googleaistudio): `PUT` `/_inference/{task_type}/{googleaistudio_inference_id}`
- [Create a Google Vertex AI inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-googlevertexai): `PUT` `/_inference/{task_type}/{googlevertexai_inference_id}`
- [Create a Hugging Face inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-hugging-face): `PUT` `/_inference/{task_type}/{huggingface_inference_id}`
- [Create an JinaAI inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-jinaai): `PUT` `/_inference/{task_type}/{jinaai_inference_id}`
- [Create a Mistral inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-mistral): `PUT` `/_inference/{task_type}/{mistral_inference_id}`
- [Create an OpenAI inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-openai): `PUT` `/_inference/{task_type}/{openai_inference_id}`
- [Create a VoyageAI inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-voyageai): `PUT` `/_inference/{task_type}/{voyageai_inference_id}`
- [Create a Watsonx inference endpoint](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-put-watsonx): `PUT` `/_inference/{task_type}/{watsonx_inference_id}`
- [Perform reranking inference on the service](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-rerank): `POST` `/_inference/rerank/{inference_id}`
- [Perform sparse embedding inference on the service](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-sparse-embedding): `POST` `/_inference/sparse_embedding/{inference_id}`
- [Perform streaming inference](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-stream-completion): `POST` `/_inference/completion/{inference_id}/_stream`
- [Perform text embedding inference on the service](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-text-embedding): `POST` `/_inference/text_embedding/{inference_id}`
- [Update an inference endpoint
](https://www.elastic.co/elastic/docs-builder/docs/4371/api/doc/elasticsearch/v8/operation/operation-inference-update): `PUT` `/_inference/{task_type}/{inference_id}/_update`, `PUT` `/_inference/{inference_id}/_update`
