8 video models · one account
AI Video Models
Browse video generation models available on Visiojoy — then try any of them with the same prompt.
Free to startOne credit balanceSwitch models anytime
Vidu Q4
ShengShu
ShengShu's character-focused model: animate a photo or combine up to 15 reference images and voice clips, up to 4K with native audio.
- Input
- Image, references (no text-only)
- Output
- 540p to 4K, up to 16s
- Best at
- Expressive characters and dialogue
Seedance 2.5
Bytedance
Longer Seedance clips with built-in audio, up to 1080p and 30 seconds per video.
- Input
- Text, image, references
- Output
- 480p to 1080p, up to 30s
- Best at
- Long, audio-led clips
Seedance 2.0
Bytedance
ByteDance's audio-video model with output up to 4K, image, video and audio references, and a cheaper Fast version.
- Input
- Text, image, references
- Output
- 480p to 4K, up to 15s
- Best at
- 4K finals with built-in audio
Seedance 2.0 Fast
Bytedance
A faster Seedance 2.0 for quick iterations, with the same audio and reference support.
- Input
- Text, image, references
- Output
- 480p or 720p, up to 15s
- Best at
- Quick drafts and iteration
Wan 3.0
Alibaba
Multimodal video generation with native audio and image, video and audio references.
- Input
- Text, image, references
- Output
- 480p to 1080p, up to 30s
- Best at
- Reference-guided video
Wan 3.0 Prime
Alibaba
A faster Wan 3.0 tier with audio and reference support, up to 1080p.
- Input
- Text, image, references
- Output
- 480p to 1080p, up to 30s
- Best at
- Fast Wan 3.0 results
MiniMax H3
MiniMax
Video up to 2K resolution with native stereo sound and multi-reference input.
- Input
- Text, image, references
- Output
- 768p or 2K, up to 15s
- Best at
- High-resolution output
Grok Imagine
xAI
xAI's video model for text prompts or reference images, up to 1080p and 30 seconds per clip, at the lowest price per second.
- Input
- Text, reference images
- Output
- 480p to 1080p, up to 30s, silent
- Best at
- Fast creative iteration
