> ## Documentation Index
> Fetch the complete documentation index at: https://docs.comfy.org/llms.txt
> Use this file to discover all available pages before exploring further.

# HeyGen Decision (Choice) - ComfyUI Built-in Node Documentation

> Pick one of your own options for the text, images, audio or video you connect, and get back the chosen option, its position and a confidence.

Pick one of your own options for the text, images, audio or video you connect, and get back the chosen option, its position and a confidence. Connect `choice` to a Branch node whose inputs carry the same option names, or route on `option_index`. A Branch node accepts at most 32 inputs, so route larger option sets with `option_index`.

## Inputs

| Parameter | Description | Data Type | Required | Range |
| - | - | - | - | - |
| `model` | Model to use. `heygen-decision-1-turbo` is faster and costs less than `heygen-decision-1` (default: `"heygen-decision-1"`). | COMBO | Yes | `"heygen-decision-1"` `"heygen-decision-1-turbo"` |
| `question` | The question the options answer, up to 5000 characters (default: "What style is the image?"). | STRING | Yes | 1 to 5000 characters |
| `options` | One option per line, used exactly as typed, up to 128 characters each. Add an option such as `other` for inputs that fit none of them (default: `photo`, `illustration`, `3d render`, `anime`, `sketch`, `other`). | STRING | Yes | 2 to 255 options |
| `images` | Images to classify, up to 4 in total, counting every frame of a batch. They are numbered `image_1` to `image_4` in order across the sockets. | IMAGE | No | Up to 4 images |
| `audio_1` | Audio to classify, up to 30 minutes. | AUDIO | No | Up to 30 minutes |
| `video_1` | Video to classify, up to 60 seconds. | VIDEO | No | Up to 60 seconds |
| `text` | Text the question is about, such as a prompt or an LLM response, up to 20000 characters. | STRING | No | Up to 20000 characters |

Give 2 to 255 options, none of them twice and none differing only in letter case, or the node raises an error. Connect at least one input to classify and name them by socket inside the question, for example "Which season is shown in image\_1?".

## Outputs

| Output Name | Description | Data Type |
| - | - | - |
| `choice` | The chosen option, exactly as it was written in `options`. | STRING |
| `option_index` | Position of the chosen option in `options`, starting at 0. | INT |
| `confidence` | Confidence in the choice, between 0 and 1. | FLOAT |

> This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! [Edit on GitHub](https://github.com/Comfy-Org/embedded-docs/blob/main/comfyui_embedded_docs/docs/HeyGenDecisionChoice/en.md)

***

**Source fingerprint (SHA-256):** `dd21b3b8021b8917e7ee4d120471974634ae6771ff17b0f49e9460f82a2fbe78`


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.