CivArchive
    Light effect transmission - High_Noise
    NSFW

    Hey everyone! This is my Light-Beam Teleport Lora that works on people, cats, and dogs, trained for Wan2.2. This model is designed to easily create that classic effect of a character (or pet) saying goodbye and being teleported away, giving you a sci-fi feeling as the subject vanishes into thin air without changing the composition.

    šŸ’»Online One-Click Generator & Workflow

    Don't want to deal with the hassle of setting up ComfyUI locally? No problem. I've set up a ready-to-use āž”ļøClick [AI Application] and a full workflow for you on the RunningHub.ai platform, so you can quickly test out what this Lora can do.

    šŸŽSpecial Bonus For You: If you're signing up for the first time, use my [link] with the invite code rh-v1182 and you'll get 1,000 free RH Coins.On top of that, you get another 100 RH Coins for logging in daily (they're valid for 24 hours). That's enough to generate plenty of images and videos!

    šŸš€Here’s how to use it in ComfyUI:

    āž”ļøClick [Comfyui] to use the full workflow and explore different effects.

    1. This is a high-noise Lora, so it should only be used in the high-noise channel withĀ Wan2.2. I didn't train a separate low-noise Lora for this because the teleport animation relies almost entirely on the high-noise Lora to work. TheĀ Wan2.2Ā low-noise model is good enough on its own to handle most of the image details.

    2. The safe range for theĀ Lora weight is between 0.5 and 1.5. I'd suggest starting atĀ 0.8Ā and tweaking from there.

    3. For theĀ sampler and scheduler, I recommend usingĀ uni_pc+beta.

    āœFor the positive prompt, use this structure:

    ymq, [Main subject description], The subject smiled and waved goodbye to the camera. Suddenly, the subject was completely enveloped by blue energy lines and rapidly decomposed into countless light particles, completely vanishing into the air.

    You just need to edit the [Main subject description] part. Simply replace it with a brief description of your subject's key features, their clothing, and the background of the image you loaded.

    šŸ’–I'm E.K., an engineer who can't draw. I'm passionate about creating more fun and practical AI models.

    If you like this model, please don't forget to drop a like, save it, and give me a follow! I'd also love to see what you create with this LoRA, so feel free to share your work in the comments. I can't wait to see everyone's creativity!

    If you run into any issues or have any suggestions, please leave a comment and let's chat. Can't wait to bring these models to life with you!

    Description

    FAQ

    Comments (4)

    mtdnicoh759Aug 18, 2025
    CivitAI

    Is there a way to achieve the reverse effect? That is, a point of light suddenly appears from nowhere and converges into a human.

    HellbornAug 18, 2025

    If not you can at least reverse the created video to get that effect.

    emilkwok
    Author
    Aug 19, 2025Ā· 1 reaction

    Yeah, that should be doable. You can try using the "first and last frame" feature in the Wan model. For the first frame, use an image that just has some points of light. For the last frame, use the image of the character after they've fully appeared. Then, you can use your prompt to control the visual transition from the first frame to the last frame.

    emilkwok
    Author
    Aug 19, 2025Ā· 1 reaction

    HellbornĀ That's definitely a cool idea, but the only catch is that playing the video in reverse might make the person's movements look a bit unnatural. Still, it's totally worth a shot

    LORA
    Wan Video 2.2 I2V-A14B

    Details

    Downloads
    606
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/18/2025
    Updated
    5/17/2026
    Deleted
    -
    Trigger Words:
    ymq

    Available On (1 platform)

    Same model published on other platforms. May have additional downloads or version variants.