Armox
    Armox Academy 📚
    CanvasVideo Nodes

    Video Nodes

    Video nodes bring motion to your creative work. Generate videos from text descriptions or animate still images using AI.

    What Are Video Nodes?

    Video Node with Kling

    A Video node using Kling V2.5 Turbo Pro for high-quality video generation.

    Video nodes generate AI videos. They can:

    • 📝 Text-to-video — Describe a scene, get a video
    • 🖼️ Image-to-video — Animate a still image
    • 🎬 Control motion — Specify camera movement and actions

    Adding a Video Node

    1. Open the node sidebar on the left
    2. Find Video in the list
    3. Drag it onto your canvas

    Video Node Inputs

    InputTypeRequiredPurpose
    PromptText (yellow)YesDescribes the video content
    Reference ImageImage (blue)NoStarting frame for I2V
    Reference VideoVideo (purple)NoStyle or motion reference
    AudioAudio (orange)NoSome models accept audio

    Video Models

    Quick & Affordable

    ModelCreditsDurationBest For
    Wan 2.5 T2V Fast605sQuick tests
    Wan 2.5 I2V Fast605sFast image animation
    Seedance 1 Lite3005sShort clips

    Standard Quality

    ModelCreditsDurationBest For
    Seedance 1 Pro Fast5005-10sFaster pro quality
    Hailuo 2.3 Fast6005sQuick smooth motion
    Wan 2.5 T2V1005sStandard text-to-video
    Wan 2.5 I2V1005sStandard image-to-video

    High Quality

    ModelCreditsDurationBest For
    Kling V2.5 Turbo Pro1,0005-10sHigh quality
    Wan 2.6 T2V1,0005-10sDetailed motion
    Wan 2.6 I2V1,0005-10sQuality animation
    Hailuo 2.31,1205-10sSmooth motion
    Kling 2.6 Pro1,2005-10sPremium quality

    Premium Quality

    ModelCreditsDurationBest For
    Veo 3.1 Fast1,5005-10sGoogle fast quality
    Sora 22,00010-20sOpenAI quality
    Veo 3.14,00010-20sHighest quality
    Sora 2 Pro4,00010-20sPremium OpenAI

    Video Models Selector

    The model selector showing all available video AI models.


    Text-to-Video Workflow

    Generate video entirely from a text description:

    Prompt Template
    [Text Node][Video Node]
    Video prompt → Generated video

    Example

    1. Add a Text Node
    2. Write a video prompt:
      A butterfly lands gently on a flower, 
      wings slowly opening and closing, 
      soft morning light, macro photography
      
    3. Add a Video Node
    4. Connect Text → Video
    5. Select model (start with Wan 2.5 Fast for testing)
    6. Click Run

    Image-to-Video Workflow

    Animate a still image:

    Prompt Template
    [Upload/Image Node][Video Node]
    Starting image → Animated video

    Example

    1. Add an Upload Node with your image
    2. Add a Video Node
    3. Connect Upload → Video (image input)
    4. Add a Text Node describing the motion:
      The woman slowly turns her head, 
      hair gently moving in the breeze, 
      soft smile forming
      
    5. Connect Text → Video (prompt input)
    6. Run to generate

    Benefits of I2V

    • ✅ Consistent subject appearance
    • ✅ Control over starting frame
    • ✅ Better for product animations
    • ✅ Use your own images/artwork

    Video Node Settings

    Duration

    Control video length:

    DurationUse Case
    5 secondsSocial media clips, loops
    10 secondsShort content, transitions
    15-20 secondsLonger narratives (premium models)

    Resolution

    Output quality:

    ResolutionBest For
    720pQuick previews, social media
    1080pStandard quality, most uses
    4KHigh-end production (if available)

    FPS (Frames Per Second)

    Motion smoothness:

    FPSEffect
    24Cinematic, film-like
    30Standard video
    60Smooth, modern look

    Additional Settings

    SettingDescription
    SeedFor reproducible results
    Motion AmountHow much movement (some models)
    Camera MotionPreset camera movements (some models)

    Writing Video Prompts

    Video prompts need to describe motion over time:

    Key Elements

    1. Subject — What's in the video
    2. Action — What happens
    3. Camera — How the camera moves
    4. Mood — Atmosphere and style

    Example Prompt Structure

    Prompt Template
    [Subject] [action/motion], 
    [camera movement], 
    [setting/environment], 
    [style/mood]

    Good Video Prompt

    A coffee cup on a wooden table, 
    steam rising and swirling gently, 
    camera slowly pushing in, 
    cozy morning light, 
    cinematic and peaceful
    

    See Also


    Processing Time

    Video generation takes longer than images:

    Model TypeTypical Time
    Fast models30 seconds - 2 minutes
    Standard models2-5 minutes
    Premium models5-10+ minutes

    💡 Tip: Start with fast models for testing, upgrade to premium for final output.


    Viewing and Downloading

    When generation completes:

    1. Video player appears in the node
    2. Click play to preview
    3. Download button to save locally
    4. Video is auto-saved to your Gallery

    Chaining Video Nodes

    Connect video outputs to other nodes:

    Video → Tool

    Prompt Template
    [Video Node][Video Tool]
    Generated video → Processed video

    Video → Video

    Prompt Template
    [Video Node 1][Video Node 2]
    First video → Extended/modified video

    Tips for Better Videos

    Prompt Tips

    • Describe motion explicitly
    • Include camera movement
    • Keep actions simple and clear
    • See Video Prompting

    Model Selection Strategy

    1. Test with fast models — Wan 2.5 Fast (60 credits)
    2. Refine your prompt — Iterate cheaply
    3. Upgrade for quality — Kling, Veo, or Sora for finals

    Quality Tips

    • Simple motions work better than complex
    • Use I2V for consistent subjects
    • Match prompt to model strengths
    • Consider adding audio separately

    Common Issues

    Video Doesn't Match Prompt

    • Simplify the motion description
    • Be more explicit about camera movement
    • Try a different model
    • Use I2V for more control

    Motion Looks Unnatural

    • Reduce motion complexity
    • Use smoother motion keywords ("gently," "slowly")
    • Try a higher-quality model
    • Check if the prompt has conflicting motions

    Subject Changes During Video

    • Use I2V with a clear starting image
    • Keep the subject description consistent
    • Use models known for consistency (Kling, Veo)

    Video Too Short

    • Check model's maximum duration
    • Use premium models for longer videos
    • Consider generating multiple clips

    Use Case Examples

    Product Animation

    Luxury perfume bottle rotating slowly on reflective surface,
    camera orbiting smoothly, 
    dramatic lighting with soft reflections,
    commercial quality
    

    Social Media Content

    Hands holding a coffee cup, steam rising,
    cozy cafe background slightly blurred,
    warm morning light,
    lifestyle video, vertical format
    

    Abstract/Artistic

    Colorful ink drops falling into water,
    swirling and mixing in slow motion,
    macro photography,
    mesmerizing and meditative
    

    Next Steps