OpenAI Conversation

The OpenAI integration adds a conversation agent powered by OpenAI in Home Assistant.

Controlling Home Assistant is done by providing the AI access to the Assist API of Home Assistant. You can control what devices and entities it can access from the exposed entities page. The AI is able to provide you information about your devices and control them.

This integration does not integrate with sentence triggers.

This integration requires an API key to use, which you can generate here.. This is a paid service, we advise you to monitor your costs in the OpenAI portal closely and configure usage limits to avoid unwanted costs associated with using the service.


To add the OpenAI Conversation integration to your Home Assistant instance, use this My button:

Generate an API Key

The OpenAI key is used to authenticate requests to the OpenAI API. To generate an API key take the following steps:

  • Log in to the OpenAI portal or sign up for an account.
  • Enable billing with a valid credit card
  • Configure usage limits.
  • Visit the API Keys page to retrieve the API key you’ll use to configure the integration.


Options for OpenAI Conversation can be set via the user interface, by taking the following steps:

  • Browse to your Home Assistant instance.
  • Go to Settings > Devices & Services.
  • If multiple instances of OpenAI Conversation are configured, choose the instance you want to configure.
  • Select the integration, then select Configure.

Instructions for the AI on how it should respond to your requests. It is written using Home Assistant Templating.

Control Home Assistant

If the model is allowed to interact with Home Assistant. It can only control or provide information about entities that are exposed to it.

Recommended settings

If enabled, the recommended model and settings are chosen.

If you choose to not use the recommended settings, you can configure the following options:


The GPT language model is used for text generation. You can find more details on the available models in the OpenAI GPT-3.5 Turbo Documentation, OpenAI GPT-4 Turbo and GPT-4 Documentation, or GPT-4o Documentation. The default is “gpt-4o”.

Maximum Tokens to Return in Response

The maximum number of words or “tokens” that the AI model should generate in its completion of the prompt. For more information, see the OpenAI Completion Documentation.


A value that determines the level of creativity and risk-taking the model should use when generating text. A higher temperature means the model is more likely to generate unexpected results, while a lower temperature results in more deterministic results. See the OpenAI Completion Documentation for more information.

Top P

An alternative to temperature, top_p determines the proportion of the most likely word choices the model should consider when generating text. A higher top_p means the model will only consider the most likely words, while a lower top_p means a wider range of words, including less likely ones, will be considered. For more information, see the OpenAI Completion API Reference.

Talking to Super Mario over the phone

You can use an OpenAI Conversation integration to talk to Super Mario and, if desired, have it control devices in your home.


Service openai_conversation.generate_image

Allows you to ask OpenAI to generate an image based on a prompt. This service populates Response Data with the requested image.

Service data attribute Optional Description Example
config_entry no Integration entry ID to use.
prompt no The text to turn into an image. Picture of a dog
size yes Size of the returned image in pixels. Must be one of 1024x1024, 1792x1024, or 1024x1792, defaults to 1024x1024. 1024x1024
quality yes The quality of the image that will be generated. hd creates images with finer details and greater consistency across the image. standard
style yes The style of the generated images. Must be one of vivid or natural. Vivid causes the model to lean towards generating hyper-real and dramatic images. Natural causes the model to produce more natural, less hyper-real looking images. vivid
service: openai_conversation.generate_image
  config_entry: abce6b8696a15e107b4bd843de722249
  prompt: "Cute picture of a dog chasing a herd of cats"
  size: 1024x1024
  quality: standard
  style: vivid
response_variable: generated_image

The response data field url will contain a URL to the generated image and revised_prompt will contain the updated prompt used.

Example using a generated image entity

The following example shows an automation that generates an image and displays it in a image template entity. The prompt uses the state of the weather entity to generate a new image of New York in the current weather state.

The resulting image entity can be used in, for example, a card on your dashboard.

The config_entry is installation specific. To get the value, make sure the integration has been installed. Then, go to Developer Tools > Services. Ensure you are in UI mode and enter the following below:

Open AI Conversation UI Mode

Select YAML Mode to reveal the config_entry value to be used in the below example automation.

Open AI Conversation YAML Mode

  - alias: "Update image when weather changes"
      - platform: state
        entity_id: weather.home
      - alias: "Ask OpenAI to generate an image"
        service: openai_conversation.generate_image
        response_variable: generated_image
          config_entry: abce6b8696a15e107b4bd843de722249
          size: "1024x1024"
          prompt: >-
            New York when the weather is {{ states("weather.home") }}"

      - alias: "Send out a manual event to update the image entity"
        event: new_weather_image
          url: '{{ generated_image.url }}'

  - trigger:
      alias: "Update image when a new weather image is generated"
      platform: event
      event_type: new_weather_image
      name: "AI generated image of New York"
      url: "{{ }}"