# <%tp.ai.chat()%>
Sends a prompt to an LLM and returns the response intot the template.
#### Command Definition
```javscript
tp.ai.chat(
promptOrMessages: string | ChatCompletionMessageParam[],
model?: string,
systemMessage?: string,
maxTokens?: number,
maxOutgoingCharacters?: number,
baseURL?: string,
apiKey?: string,
organization?: string
)
```
#### Command arguments
- `promptOrMessages`: The prompt to send to the large language model. if it is a string, a single prompt is sent. For more advanced chat completions, use the [ChatCompletionMessageParam](https://platform.openai.com/docs/api-reference/chat/create) format.
- `model`: (optional) name of the large language model, like gpt-4-turbo-preview or gpt-3.5-turbo
- `systemMessage`: (optional) Overrides the system message as defined in settings.
- `maxTokens`: (optional) Overrides the max tokens as defined in settings.
- `maxOutgoingCharacters`: (optional) Overrides the max characters that can be sent to the LLM, as defined in settings.
- `baseURL`: (optional) Overrides the base URL as defined in settings.
- `apiKey`: (optional) Overrides the API key as defined in settings.
- `organization`: (optional) Overrides the organization as defined in settings.
#### Example
`<%tp.ai.chat("How old is the moon?")%>`
---
# <%tp.ai.selection_or_nearest()%>
Grabs the currently selected text in a document, or selects the line where the cursor is and returns the entire line.
#### Command Definition
```javscript
tp.ai.selection_or_nearest()
```
#### Example
`<%tp.ai.selection_or_nearest()%>`
---
# <%tp.ai.content_without_properties()%>
Similar to [Templater's tp.file.content](https://silentvoid13.github.io/Templater/internal-functions/internal-modules/file-module.html#tpfilecontent), but does not return any of the properties.
#### Command Definition
```javscript
tp.ai.content_without_properties()
```
#### Example
`<%tp.ai.content_without_properties()%>`
---
![[Examples Index]]