Creators and Presenters
Develop talking-head concepts with a clear voice, expressive delivery, and a consistent visual persona.
HeyGen Avatar 6 can animate a photo with speech, matching facial expression and gestures to the delivery. It supports motion direction and 1080p or 2K output for talking presenters and character scenes.
Explore six official Avatar 6 clips, then use an editable starting prompt to create your own scene. Starting prompts describe the scene; they do not reproduce the original identity or performance.
HeyGen Avatar 6
An excited presenter
HeyGen Avatar 6
A surprised reaction on the couch
HeyGen Avatar 6
A celebration with raised fists
HeyGen Avatar 6
A conversation with varied gestures
HeyGen Avatar 6
A smiling presenter at a desk
HeyGen Avatar 6
Two speakers over coffee
A talking-avatar model that connects speech, expression, and body movement.
HeyGen released Avatar 6 on November 17, 2026. It animates a reference photo or video using speech audio, with movements shaped by the content and delivery rather than a repeated gesture loop.
A photo provides a visual starting point; an eligible video reference can also guide personal mannerisms. Realistic human video references have consent and access requirements. The model supports 1080p and 2K output, with video length following the supplied audio.
HeyGen Avatar 6 combines three kinds of direction:
Use only images, voices, and likenesses you have permission to use. Realistic personal avatars require the applicable identity and consent checks.
Output dimensions depend on the workflow. 2K is a resolution option, rather than one fixed pixel size for every aspect ratio.
Six capabilities help turn a visual reference and speech into a directed performance.
Speech audio guides mouth movements and facial delivery, keeping a talking presenter connected to what is being said.
Gestures can follow the meaning and emphasis of the script, including pointing, counting, and directional movement.
The model generates motion around the performance instead of relying on the same short gesture loop.
Describe expression, posture, gestures, and camera direction. A calmer motion setting can reduce gesture size and frequency.
Choose between documented 1080p and 2K output modes. The resulting dimensions depend on the reference shape and workflow.
HeyGen’s official showcase includes two avatars sharing one frame and a conversation. This is a distinct workflow from a single-presenter animation.
Talking presenters can help communicate a message across creative, learning, and business settings.
Develop talking-head concepts with a clear voice, expressive delivery, and a consistent visual persona.
Plan lessons, onboarding explainers, and step-by-step demonstrations with gestures that support the spoken message.
Explore spokesperson scenes, product introductions, and internal updates using permitted identities and assets.
Experiment with character expression, mood, and conversational staging before developing a finished scene.
Use an example as a starting point, refine your brief, and review the generator’s controls.
Apply an example or describe a presenter, setting, and delivery. Use an original character or a likeness you have permission to use.
Review the selected model and choose its aspect ratio, resolution, and duration. Keep the scene and dialogue suited to the available settings.
Submit when ready, review the result, and adjust the prompt or framing. Applying an example only fills the generator; it does not start generation.
Verified Avatar 6 model capabilities; exact output dimensions depend on the reference and workflow.
| Developer | HeyGen |
| Public release | November 17, 2026 |
| Core task | Speech-driven avatar video generation |
| Reference media | One image, one eligible video, or both |
| Speech input | Speech audio; script and voice in avatar workflows |
| Resolution options | 1080p and 2K |
| Duration | Follows speech audio length |
| Motion direction | Natural-language gesture, expression, posture, and camera instructions |
| Expressiveness | Default or reduced motion |
| Output framing | Reference shape or workflow-specific aspect ratio |
| Human likeness | Permission required; realistic video references have additional consent and access requirements |
Last updated: November 18, 2026
Compare the current model with its predecessor using documented performance controls and official product descriptions.
| Specification | HeyGen Avatar 6 | HeyGen Avatar V |
|---|---|---|
| Performance approach | Speech-context-aware gestures and varied movement | Reference-driven animation and motion |
| Motion prompts | Gestures, expression, posture, and camera direction | Body motion and hand gestures; eligible reference required |
| Reduced-motion setting | Default or reduced expressiveness | Not listed in the Avatar V guide |
| Visual references | Photo or eligible video reference | Eligible digital twin, studio avatar, or photo avatar looks |
| Output resolution | 1080p or 2K | 1080p shown in the documented example; full range not listed |
HeyGen Avatar 6 Advantages: Speech-aware motion, explicit calmer delivery, and documented 2K output provide more direction for a new presenter performance.
HeyGen Avatar V Advantages: Reference-driven animation supports eligible existing avatar looks and includes an optional natural-blinking control.
Explore Avatar 6 for script-aware gestures and higher-resolution output. Consider Avatar V when maintaining an established reference-driven avatar workflow. Compare both with your own permitted inputs.
Choose a Picir AI plan for video creation. The generator displays the actual selected model and the credit cost for its current settings.
Answers about speech, references, output, examples, and creating presenter scenes.
Explore the HeyGen Avatar 6 AI Video Generator page on Picir AI. Browse official clips and edit a starting prompt in the video generator.
Avatar 6 is HeyGen’s speech-driven avatar model, publicly released on November 17, 2026. It turns a photo or eligible video reference and speech into a talking performance with facial expression and body movement.
Avatar 6 is designed to match gestures to speech context and vary motion across a performance. It also documents 1080p and 2K outputs and a reduced-motion option. Avatar V uses reference-driven animation for eligible avatar looks.
Yes. Avatar 6 accepts natural-language directions for gestures, expression, posture, and camera behavior. Its reduced-motion setting asks for fewer, smaller gestures and calmer expressions.
A photo can be enough to establish the subject. Eligible video footage can additionally guide mannerisms. Realistic human video references have additional consent and access requirements; use assets you have permission to use.
The documented resolution options are 1080p and 2K. Video length follows the supplied speech audio, and pixel dimensions depend on the reference shape and workflow. Review the selected generator model for the controls available in your current creation.
HeyGen’s official showcase demonstrates two avatars in a shared conversation. That is a separate conversation workflow; a single-presenter request does not automatically create a two-person scene.
The clips are official Avatar 6 examples. Their buttons contain newly written starting prompts inspired by the visible scene, using original characters rather than reproducing the presenters’ identities. They do not supply the original voice, audio, or full production settings.
Use the Picir AI plans below and check the credit estimate in the generator before submitting. The estimate follows the selected model and settings. Applying an example does not submit a generation or spend credits.
Describe the subject, setting, framing, and one clear delivery style. Keep dialogue brief and gestures specific. Use an original character or obtain permission for any real person’s likeness and voice.