CivArchive
    [Workflow] LTX 2.3 - Retake Video - v1.0
    NSFW
    Preview 138608818
    Preview 138608849

    Re-shoot one bad segment of a video without redoing the whole clip

    Open this ready-to-run workflow on Floyo. No install needed.

    HOW IT WORKS

    Step 1. Upload your video. The clip with one section you want to fix. LTX 2.3 reads the surrounding footage as context, so the retake matches the style and motion around it. Works great with: generated clips · talking heads · character shots · product footage

    Step 2. Describe the replacement. Write a prompt for what the segment should look like, specific about the action, camera angle, and subject behavior. "The man walks forward on the street and looks around casually with subtle body motion" gives the model a clear target.

    Step 3. Set the window. Enter the start time in seconds where the problem begins, and the duration to cover. Default is 5 seconds. Keep it tight around the bad moment, since a narrow window blends better with the footage on both sides.

    Step 4. Pick a retake mode and run. Choose whether to replace the video, the audio, or both. LTX 2.3 regenerates that window and stitches it back into the clip. Ready for: Premiere · DaVinci Resolve · After Effects · any editor

    First time? Leave the defaults. Set your start time, keep the duration around 5 seconds, and pick a mode. The rest is handled.

    Overview

    This workflow re-generates one segment of an existing video using LTX Video 2.3, without rerunning the whole clip. You mark where the problem starts and how long it runs, describe the fixed version, and the model regenerates only that window and blends it back in. It reads the footage before the retake point as context, so the replacement matches the visual style and motion instead of jumping. A retake mode controls what changes: replace_video_only fixes the visuals and keeps the original audio, replace_audio_only fixes the sound and keeps the picture, and replace_audio_and_video redoes both. It is a surgical fix, not a whole-clip transform. You upload a clip, set the window, write a prompt, and get the patched video back in about 70 seconds. No setup, no nodes to wire.

    Who it's for: editors and video creators who want to fix one bad moment in a generated clip in ComfyUI without regenerating the whole thing. Not for: full-length restyling or structural changes across the clip. A full video-to-video workflow is the better tool once the change spans the whole composition.

    Why Floyo

    Floyo is the only ComfyUI platform built for teams in the browser.

    • Made for teams. Share run history, files, and models across your whole team. A teammate opens your exact run and picks up where you left off. No file handoffs, no version confusion.

    • No install, no setup. Every workflow and model is preloaded. Open it in your browser and run. Nothing to download, nothing to configure.

    • No local hardware. Workflows run on H100 NVL GPUs, so heavy models run fast without a card of your own. Your VRAM stops being the limit.

    • Open and closed models in one place. Floyo runs open-source workflows and API models side by side.

    How to use (in your browser on Floyo)

    1. Upload the clip you want to fix.

    2. Write a prompt for the replacement, set the start time and duration, and pick a retake mode. Then run.

    3. Download the patched clip, ready for any editor.

    Expectations The models are preloaded, so there is nothing to download. Running it needs a free Floyo account. Each retake comes back in about 70 seconds, far quicker than regenerating the full clip. A few things decide the result. Keep the window tight around the bad moment, since a narrow retake stays more coherent with the footage on both sides. Clips with steady lighting and motion across the retake window blend the most naturally. Match your prompt to what the rest of the clip is already doing, so the patched segment does not stand out.

    Use Cases

    Fix Motion Glitches. Point retake at a 3 to 4 second window where the motion broke and regenerate just that part.

    Targeted Reshoots. Replace a gesture that reads too fast, a look that misses, or a camera move that breaks the flow, without touching the rest.

    Audio-Only Repairs. Keep good footage and regenerate just the sound for a segment where the audio is off.

    Salvage a Near-Miss. Rescue a clip that came out well except for one bad moment, instead of rolling the dice on a full rerun.

    What can LTX 2.3 retake fix?

    Description

    Workflows
    LTXV

    Details

    Downloads
    52
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/3/2026
    Updated
    8/7/2026
    Deleted
    -

    Files

    WorkflowLTX23Retake_v10.json

    Mirrors

    CivitAI (1 mirrors)