Armox
    Armox Academy 📚
    KI-Modelle ReferenzVideo-Modelle

    Video Models

    Video Models in Armox transformieren Textbeschreibungen und Images in dynamischen Video-Content. Von kurzen Clips bis zu cineastischen Produktionen — diese Modelle bringen Ihre Ideen in Bewegung.

    Overview

    Video Models können:

    • Text-to-video — Videos aus Textbeschreibungen generieren
    • Image-to-video — Still Images animieren
    • Motion control — Bewegungen und Kameraarbeit gezielt steuern
    • Audio generation — Einige Models enthalten native Audio
    • Multi-shot editing — ZusammenhĂ€ngende Videosequenzen erstellen

    Available Video Models

    ModelProviderCostDurationBest For
    Kling V2.5 Turbo ProKuaishou1,000 credits5-10sCinematic quality
    Kling 2.6 ProKuaishou1,200 credits5-10sNative audio
    Veo 3.1Google4,000 credits4-8sPremium quality
    Veo 3.1 FastGoogle1,500 credits4-8sFast generation
    Seedance 1 ProByteDance1,000 credits2-12sLong videos
    Seedance 1 Pro FastByteDance500 credits2-12sBudget option
    Seedance 1 LiteByteDance300 credits2-10sDance/motion
    Wan 2.6 I2VWan1,000 credits5-15sAudio sync
    Wan 2.6 T2VWan1,000 credits5-15sText-to-video
    Sora 2 ProOpenAI4,000 credits4-12sPremium quality
    Sora 2OpenAI2,000 credits4-12sGood quality
    Hailuo 2.3MiniMax1,120 credits6-10sPrompt optimization
    Hailuo 2.3 FastMiniMax600 credits6-10sFast generation
    Video-01 DirectorMiniMax1,000 creditsVariableCamera control
    Gen4 AlephRunway1,000 credits5sVideo editing

    Connection Colors

    Im Armox Canvas verwenden Video-Connections green Handles und Edges:

    • Input Handle: Green circle auf der linken Seite von Nodes
    • Output Handle: Green circle auf der rechten Seite von Nodes
    • Connection Edge: Green line, die Nodes verbindet

    Common Settings

    Aspect Ratio

    • 16:9 — Landscape (YouTube, presentations)
    • 9:16 — Portrait (TikTok, Reels, Stories)
    • 1:1 — Square (Instagram feed)

    Duration

    Die meisten Models unterstĂŒtzen 5–10 Sekunden Clips. Einige bieten lĂ€ngere Durations.

    Resolution

    • 720p — Standard quality, schneller
    • 1080p — High quality, empfohlen
    • 4K — Maximum quality (bei manchen Models)

    Choosing the Right Model

    For Maximum Quality

    • Veo 3.1 (4,000 credits) — Googles bestes
    • Sora 2 Pro (4,000 credits) — OpenAI Premium
    • Kling 2.6 Pro (1,200 credits) — Exzellent inklusive Audio

    For Speed and Value

    • Seedance 1 Pro Fast (500 credits) — Gute Balance
    • Hailuo 2.3 Fast (600 credits) — Schnelle Results
    • Seedance 1 Lite (300 credits) — Budget option

    For Audio

    • Kling 2.6 Pro — Native audio generation
    • Veo 3.1 — Audio support
    • Wan 2.6 models — Audio synchronization

    For Image Animation

    • Wan 2.6 I2V — Image-to-video specialist
    • Kling models — Starke Image Animation
    • Seedance models — Gute Motion Consistency

    Best Practices

    1. Start with shorter durations — Erst Konzepte testen
    2. Use negative prompts — Ungewollte Elemente vermeiden
    3. Describe motion — Bewegung konkret beschreiben
    4. Consider audio needs — Models mit Audio wĂ€hlen, wenn nötig
    5. Match aspect ratio to platform — 9:16 fĂŒr Social, 16:9 fĂŒr YouTube

    Next Steps

    Explore individual model documentation for detailed settings and use cases.