Which AI video models copy a performance from a video?
Runway's Act-One drove a character from one performance video in October 2024 and Act-Two tracked head, face, body and hands in July 2025. Luma's Modify Video, Wan 2.2 Animate, Kling's Motion Control from December 2025 and PixVerse's Mimic in February 2026 carry the same idea. From 6 source pages, read 2026-09-25 to 2026-10-25; no maker pays to be listed.
| Date | Line | Version | What changed |
|---|---|---|---|
| 25 October 2024 | Runway | Act-One | Act-One drives a character from one performance video and one image |
| 6 December 2024 | Runway | Act-One | Act-One puts a performance onto characters inside existing videos |
| 2 June 2025 | Luma Ray | Not named in the note | Modify Video keeps a performance and changes the world around it |
| 17 July 2025 | Runway | Act-Two | Act-Two tracks head, face, body and hands |
| 10 November 2025 | Wan | Wan 2.2 | Wan 2.2 Animate moves a character photo with the performance from a reference video |
| 17 December 2025 | Kling AI | Kling 2.6 | Kling 2.6 Motion Control copies an action and expression, up to 30 seconds in one take |
| 22 December 2025 | Kling AI | Not named in the note | Motion Control in the API takes a reference image and a reference video together |
| 28 February 2026 | PixVerse | Not named in the note | Mimic motion control is released |
| 4 March 2026 | Kling AI | Kling 3.0 | Kling 3.0 Motion Control launches |
Inclusion rule. Entries whose note bears directly on the question. Order. By date, oldest first; undated rows last.
Performance transfer takes the movement from one video and puts it on another character, so a person can act a scene once and have it played by someone designed in an image. It is one of the few threads in the register that runs from 2024 to 2026 across several makers with steady additions.
The notes add inputs over time. Act-One took a driving video and a character image; in December 2024 it could put a performance onto characters inside existing videos. Kling's Motion Control API needs a reference image and a reference video together, and its 3.0 version followed in March 2026.
1In the makers' words
Runway, 25 October 2024
A new way to generate expressive character performances inside Gen-3 Alpha using a single driving video and character image.
Runway, changelog
- Act-One in Gen-3 AlphaAct-One drives a character from one performance video and one image
Runway, 6 December 2024
Act-One now allows you to transpose your performances directly onto characters inside of existing videos.
Runway, changelog
- Character reference videos in Act-OneAct-One puts a performance onto characters inside existing videos
Luma Ray, 2 June 2025
With Modify Video, you can keep what matters and evolve everything else.
Luma, Modify Video post
- Modify VideoModify Video keeps a performance and changes the world around it
Runway, 17 July 2025
Next-generation motion capture model with major improvements in generation quality and support for head, face, body and hand tracking.
Runway, changelog
- Act-TwoAct-Two tracks head, face, body and hands
Wan, 10 November 2025
Users simply upload a character photo and a reference performance video, and the model transfers the expressions and actions from the video onto the character in the image
Alibaba Cloud Model Studio, newly released models
- wan2.2-animate-moveWan 2.2 Animate moves a character photo with the performance from a reference video
Kling AI, 17 December 2025
30 seconds of high-difficulty action in one seamless take, with total command over movement, expressions, lip-sync, and gestures.
Kling AI, release notes
- 2025-12-17Kling 2.6 Motion Control copies an action and expression, up to 30 seconds in one take
Kling AI, 22 December 2025
both reference images and reference videos need to be uploaded simultaneously
Kling AI, API updates
- 12/22/2025Motion Control in the API takes a reference image and a reference video together
PixVerse, 28 February 2026
Added the Motion Control (Mimic) endpoint.
PixVerse, API changelog
- 2026-02-28Mimic motion control is released
Kling AI, 4 March 2026
V3.0 Motion Control Launched
Kling AI, API updates
- 03/04/2026Kling 3.0 Motion Control launches
Motion and performance · Which AI video models let you set or clone a voice? · Which AI video models add HDR or higher frame rates?
2Notes read
- Runway, changelog, read 2026-09-25
- Luma, Modify Video post, read 2026-09-25
- Alibaba Cloud Model Studio, newly released models, read 2026-09-25
- Kling AI, release notes, read 2026-09-25
- Kling AI, API updates, read 2026-09-25
- PixVerse, API changelog, read 2026-09-25
3Where new versions end up
Release notes such as “Which AI video models copy a performance from a video?” matter once a pipeline runs the models; SceneMixer's guides describe how it does:
Doubao Seed 2.1 Pro is the family’s full-size tier at 100% of the base text rate, twice the default. Same three jobs — reading, writing, storyboards — with more headroom on long, crowded episodes: it holds the storyboard’s strict format more reliably and keeps track of many characters and locations in one pass.
SceneMixer, Doubao Seed 2.1 Pro (read 2026-10-13)
Recommended is the badge on the tier the site suggests for most character sheets; the default for new accounts stays the cheaper tier.
SceneMixer, Qwen-Image 3.0 Pro: the recommended tier — “Is Recommended the same as default?” (read 2026-10-13)