HeyGen Avatar 6

HeyGen's newest talking avatar model, run on HeyGen's API. Wire a photo of a person (or a short clip) and a speech track, and get a lip-synced video of them saying it, as long as the audio. 1080p or 2K, with an optional motion prompt for gestures and expression.

Try HeyGen Avatar 6 in the playground

Last updated View as MarkdownAgent setup

HeyGen Avatar 6

Badge: New
Node type: video:heygen_avatar_6
Category: video

Description

HeyGen's newest talking avatar model, run on HeyGen's API. Wire a photo of a person (or a short clip) and a speech track, and get a lip-synced video of them saying it, as long as the audio. 1080p or 2K, with an optional motion prompt for gestures and expression.

Pricing

  • Cost: ~15 credits per second
  • Per-second rate: The figure is the rate for a 5-second clip. Shorter clips can cost a bit more per second. You're always charged for the real length.

Canvas ports

These appear as port handles on the left side of the node.

ID Label Details
audio_url Audio AUDIO (required). The speech the avatar performs. The video runs the length of this audio (up to 60 seconds here). For a script, wire a HeyGen TTS node in.
reference_image_url Reference Image IMAGE. A photo of the person, used with their permission. Wire this, a Reference Video, or both. The output keeps its shape.
reference_video_url Reference Video VIDEO. Optional clip of the person's own gestures, at least 2 seconds. HeyGen takes clips of real people only from accounts with video-reference access; animated or illustrated characters work everywhere. Add a matching photo for the best likeness.

These render as form fields in the right-side config panel when the node is selected.

ID Label Details
resolution Resolution TEXT · options: 1080p, 2k · default: "1080p". The short edge is 1080 px at 1080p and 1620 px at 2K. The shape follows the reference.
motion_prompt Motion Prompt TEXT · multiline. Optional direction for gestures, expression, posture and camera, e.g. "warm smile, small open-handed gestures". Up to 1,200 characters. Wins over Expressiveness where they overlap.
motion_expressiveness Expressiveness TEXT · options: default, less · default: "default"

Outputs

ID Label Type
video Generated Video VIDEO

How to wire it

Avatar 6 animates a person you give it with speech you give it. Two wires:

  • Audio (required): the speech. The video runs exactly as long as this track. To start from a script, wire a HeyGen TTS node in first.
  • Reference Image, Reference Video, or both: who is talking. A clear, front-facing photo works in every account. A clip carries the person's own gestures and gives the closest likeness, and a matching photo next to it helps.

There is no avatar to set up first and no voice picker on this node.

Who you can animate

  • A photo of a real person works, used with their permission.
  • A clip of a real person needs video-reference access on HeyGen's side. Without it HeyGen refuses the run and your credits come back. Clips of animated or illustrated characters work everywhere.
  • A clip has to be at least 2 seconds long.

What it costs

Billed per second of audio, measured from the file on the server before any credits move.

Resolution Credits per second 10 s of audio 30 s of audio
1080p (default) 15 150 450
2K 21.75 218 653

Anything under 3 seconds bills as 3. Audio over 60 seconds is refused before you are charged, never cut short: split the script into takes and join them in the Video Editor. If the file's length cannot be read, the run is refused and nothing is charged.

The number on the canvas is the most a run can cost (up to 60 seconds at the resolution you picked), and a run needs that much in your balance to start, even for a short clip. You are charged only for the measured length.

Shape and motion

The video keeps the shape of the reference. At 1080p the short edge is 1080 px (1920 x 1080 for a 16:9 photo), at 2K it is 1620 px (2880 x 1620). Crop the photo first if you want a different shape.

Motion Prompt directs gestures, expression, posture and camera in plain words, for example "warm smile, small open-handed gestures, looks into the lens". Expressiveness set to Less gives fewer, smaller gestures and a calmer face. Where the two disagree, the motion prompt wins.


Auto-generated from the Wireflow node registry.

For AI agents: the full documentation index is at /llms.txt, and most docs pages are available as Markdown by adding .md to the URL.

© 2026 Wireflow. All rights reserved.