# Motion Transfer > Make a still character perform the moves of a reference video, such as a dance or a gesture, with Kling 2.6, Kling 3.0 or Wan Animate. Source: https://nodaro.ai/docs/nodes/video/motion-transfer The **Motion Transfer** node makes a still character perform the moves of a reference video. Connect an image of the character and a video of the motion, for example a dance, and the node returns a video of your character doing those moves. You choose between Kling and Wan Animate models. - Found in: Video › Animate & Perform - Output: video - API type: `motion-transfer` ## When to use it - Make an AI-generated character dance or perform a move from real footage. - Animate an illustrated character with the movements of a real person. - Make consistent character videos from a single image. - Replace a person in a clip with your own character, with Wan Animate Replace. ## Quick start ### Add the node Press Tab on the canvas and choose **Video › Animate & Perform › Motion Transfer**. ### Connect the character and the motion Connect a clear, full-body image of the character to **Character**. Connect the reference video to **Motion**, for example from [Upload Video](https://nodaro.ai/docs/nodes/video/upload-video). ### Choose the model Open the settings panel and choose a model under **Provider**, then the **Resolution**. Optionally describe the motion in the **Prompt (Optional)** field. ### Run it Click **Run**. The video appears on the node. Workflow: A generated character performs a dance from an uploaded clip, trimmed to the moves that matter. - Upload Video → Trim Video - Generate Image → Motion Transfer (character) - Trim Video → Motion Transfer (motion) ## Inputs | Input | Accepts | What it does | | --- | --- | --- | | **Character** | Image nodes | The still image of the character. Required. | | **Motion** | Video nodes | The reference video with the movement. Required. | | **Prompt** | Text nodes | A description of the motion or the scene. Optional. | | **Negative** | Text nodes | What to avoid. Optional. | | **Assets** | Character, Location, Object, Animal/Creature and Create Face nodes | Identities you can mention with `@` in the prompt. | The output, **Video**, is the URL of the generated video. ## Settings | Setting | What it does | | --- | --- | | **Provider** | The model: **Kling 2.6** (the default), **Kling 3.0**, **Wan Animate Move** or **Wan Animate Replace**. | | **Prompt (Optional)** | A description of the motion. | | **Negative Prompt (Optional)** | What to avoid. | | **Character Orientation** | Kling models only. **Image** keeps the character as in the picture, up to 10 seconds on Kling 2.6. **Video** follows the orientation in the video, up to 30 seconds on Kling 2.6. | | **Background Source** | Kling 3.0 only. **Input Video** keeps the scene of the reference video. **Input Image** uses the background of your image. | | **Resolution** | 720p or 1080p on Kling models, 720p by default. 480p, 580p or 720p on Wan Animate, 480p by default. | | **Pre & post text** | Text always added before and after the prompt. See [Prompt pre and post text](https://nodaro.ai/docs/concepts/prompt-pre-post-text). | The node detects the length of the connected video and shows it, because the cost follows the length. ## Models | Model | What it does | | --- | --- | | **Kling 2.6** | Standard motion transfer, 720p or 1080p, from 5 to 30 seconds. | | **Kling 3.0** | Premium motion control with prompts, and a choice of background source. | | **Wan Animate Move** | Moves the character from your image within the scene of the video. Lighter and faster. | | **Wan Animate Replace** | Replaces the character in the video with the character from your image. Lighter and faster. | | Model | Maker | Modes | Credits | Details | | --- | --- | --- | --- | --- | | [Kling 2.6 Motion Transfer](https://nodaro.ai/docs/models/video/kling-2-6-motion-transfer) | Kuaishou | Motion transfer | from 80 | Transfer the motion from a driving video onto a still subject. Kling 2.6 base. | | [Kling 3.0 Motion Transfer](https://nodaro.ai/docs/models/video/kling-3-0-motion-transfer) | Kuaishou | Motion transfer | from 150 | Premium motion transfer via Kling 3.0. | ## Credits The price depends on the model, the resolution and the length of the reference video. - **Kling 2.6** at 720p: 80 credits for 5 seconds, 150 for 10 seconds, 230 for 15 seconds and 450 for 30 seconds. At 1080p, 10 seconds cost 230 credits. - **Kling 3.0** at 720p: 150 credits for 5 seconds and 300 for 10 seconds. At 1080p, 10 seconds cost 500 credits. ## Tips - **Use a full-body image.** A clear, full-body character works best, especially for dance. - **Choose clean motion.** A reference video with distinct, clearly visible movements transfers best. - **Trim the reference.** Cut the video to the moves you need with [Trim Video](https://nodaro.ai/docs/nodes/video/trim-video); a shorter clip also costs less. - **Test on Wan Animate.** It is faster, so try the motion there before you move to Kling. - **Choose the background on Kling 3.0.** Use **Input Image** for a clean background from your picture, or **Input Video** to keep the reference scene. ## Frequently asked questions ### What does Motion Transfer need? A still image of the character and a reference video with the motion to copy. A prompt is optional and describes the motion or the scene. ### Which Motion Transfer model should I choose? Kling 2.6 is the standard, lower-cost choice for clips of 5 to 30 seconds. Kling 3.0 is the premium choice and lets you choose whether the background comes from the video or the image. Wan Animate Move and Wan Animate Replace are lighter and faster, good for tests. ### How many credits does Motion Transfer cost? The price depends on the model, the resolution and the length of the reference video. Kling 2.6 costs 80 credits for 5 seconds and 150 for 10 seconds at 720p; Kling 3.0 costs 150 and 300. The node detects the video's length and shows the cost. ### How long can the result be? The length follows the reference video. On Kling 2.6, Character Orientation set to Image allows up to 10 seconds, and set to Video up to 30 seconds.