> For the complete documentation index, see [llms.txt](https://xfigura.gitbook.io/xfigura-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://xfigura.gitbook.io/xfigura-docs/models/video-models.md).

# Video Models

<table><thead><tr><th width="161.33331298828125">Models</th><th width="97.666748046875">Credits</th><th>Description</th><th>xFigura Tip</th></tr></thead><tbody><tr><td>Runway</td><td>8/s</td><td>A text-to-video model focused on creative flexibility, motion design, and visual storytelling across stylized and experimental aesthetics.</td><td>Use Runway when exploring visual ideas, animated concepts, or stylized narratives. It’s ideal for motion-first experimentation where art direction matters more than physical realism.</td></tr><tr><td>Kling AI</td><td>3/s</td><td>A family of text-to-video and image-to-video models built for creative control, from lightweight generation to film-grade outputs with realistic motion and lighting.</td><td>Choose lighter Kling 2.1 versions for quick ideation and experimentation. Use the more advanced versions (Pro anf Master) when precise camera movement, realistic motion, and cinematic polish matter most.</td></tr><tr><td>Sora 2</td><td>4/s</td><td>OpenAI's flagship video model with improved physics, multi-shot consistency, and synchronized dialogue and sound effects generated from a single prompt.</td><td>Use for realistic, physics-driven video with audio - ads, pre-vis, and cinematic storytelling.</td></tr><tr><td>Bytedance Seedance</td><td>4/s</td><td>A fast, multi-modal video model converting text or images into fluid, coherent video with strong motion continuity and scene pacing.</td><td>Use Seedance when you need fast video outputs with smooth motion and natural transitions. It’s a strong choice for quick concepts, short clips, and motion-focused previews.</td></tr><tr><td>Bytedance Seedance 2.0</td><td>8/s</td><td>ByteDance's cinematic video model with multimodal input, native audio, multilingual dialogue, and strong character consistency across multi-shot sequences.</td><td>Use for reference-guided, story-led video where audio, consistency, and camera control all matter.</td></tr><tr><td>Runway Gen-4 Aleph</td><td>8/s</td><td><strong>Video to Video</strong> Transforms existing footage by applying new styles and aesthetics via text prompt, preserving the original structure and motion.</td><td>Use to restyle footage without reshooting -changing look, aesthetic, or treatment in one pass.</td></tr><tr><td>Google Veo 3.0</td><td>5/s</td><td>Google DeepMind's cinematic video model with native audio - dialogue, sound effects, and ambience generated alongside visuals in a single pass.</td><td>Use when your video needs audio built in from the start.</td></tr><tr><td>Google Veo 2.0</td><td>5/s</td><td>A cinematic video model family that progressively improves visual realism, motion quality, and audio - with newer versions generating synchronized dialogue and sound effects directly from text prompts.</td><td>Use earlier Veo 2.0 versions for fast visual previs and concept shots. Use the Veo 3.0 when you need film-like realism with sound, dialogue, and polished cinematic timing.</td></tr><tr><td>Topaz Upscaler</td><td>20/s</td><td>A professional video upscaler that enhances resolution, stabilizes footage, reduces noise, and refines textures for a clean, production-ready finish.</td><td>Use Topaz when quality matters most. It’s best for final exports, restoration workflows, or any video that needs maximum clarity, stability, and visual fidelity.</td></tr><tr><td>Fal AI Video Upscaler</td><td>25/s</td><td>A video upscaler that improves sharpness and clarity frame by frame while preserving the original structure, motion, and creative intent.</td><td>Use this when you want a quick, reliable quality boost for AI-generated or low-resolution videos without changing the look or motion. Ideal as a final polish step before delivery.</td></tr></tbody></table>