Zopia × SVAiFF
2026 Los Angeles–Silicon Valley AI Film Festival
Create with AI. Be seen by the world.
Submission deadline ·
200 Zopia Universal Credits + 800 Vidu Credits
Credit eligibility and redemption follow the campaign terms.
View event
Vidu Q4 Preview · Now on Zopia
Vidu Q4 — AI Video Generator
Bring your characters into the stories you want to tell.
Preparing your creative tools…
From references to stories with sound
Characters, scenes and voices can all be your starting point.
Vidu Q4 is a video generation model from Shengshu Technology, supporting reference-to-video and image-to-video. Combine images to define characters, environments and visual style, or start with a single image and describe what happens next. Add voice references to bring both your characters’ appearance and their voices into the story.
What you can create with Vidu Q4
Multiple image references · Characters and scenes
Bring different references into the same story.
Reference-to-video supports 1–15 images. Combine character designs, scene compositions and style references, then explain their relationships in your prompt. From a single protagonist to a fantasy world, give your ideas a concrete foundation.
Image-to-video · Action and camera movement
The image sets the scene. You describe what comes next.
Upload a starting frame and describe the subject’s actions, camera direction and changes in the scene to create a 3–16 second video. Guide running, chasing, smoke and lighting with specific descriptions of movement.
Voice references · Multi-shot storytelling
Give your characters a voice and your scenes a next chapter.
Add up to 3 voice references in reference-to-video mode. Describe dialogue and emotion in your prompt to help shape each character. With up to 16 seconds per generation, there is room for entrances, action and shot changes.
Selected official Vidu examples
Watch a scene. Imagine your next one.
These videos from the Vidu Q4 website showcase different creative directions. Find inspiration in character relationships, atmosphere and the rhythm of movement.
Reference-to-video
Build a world for your characters
Characters and fantasy settings meet on screen, giving your character designs a world to explore.
Image-to-video
Let the action guide the camera
See how character movement, camera changes and environmental details work together in dynamic scenes.
Showcase
Explore more storytelling styles
Find your direction in scenes from different genres, then use references to express your own ideas.
Turn an idea into a video in three steps
01
Prepare your references
Upload one starting frame for image-to-video. For reference-to-video, combine character, scene and style images, with optional voice references.
02
Describe what happens next
Specify the subject, actions and camera changes. For dialogue, describe who speaks, what they say and the emotion you want to convey.
03
Generate and refine
Choose a duration and submit. View progress and results in the creative canvas, then refine your prompt or references to develop the scene.
Know your creative options before you start
Choose an input mode that suits your idea and organize your references and prompt clearly.
| Current model | Vidu Q4 Preview |
|---|---|
| Creation modes | Reference-to-video and image-to-video |
| Video duration | 3–16 seconds |
| Image references | 1–15 images for reference-to-video; 1 starting frame for image-to-video |
| Voice references | Optional 0–3 audio clips in reference-to-video mode |
| Audio generation | Synchronized audio and video; enable sound as needed |
Image-to-video does not use an end frame or voice references. Both modes require image input.
More about Vidu Q4
Can Vidu Q4 generate a video from text alone?
The currently integrated Vidu Q4 Preview supports image-to-video and reference-to-video. Both require images. Prepare a starting frame or reference images, then describe the action, scene and camera in text.
Which mode should I choose?
Choose image-to-video if you have an image you want to use as the starting frame. Choose reference-to-video to combine characters, scenes and styles or add voice references. Explain the purpose of each reference in your prompt.
How many references can I add?
Reference-to-video supports 1–15 images and an optional 0–3 voice references. Image-to-video uses 1 starting frame and does not support an end frame or voice references. Video references are not currently supported.
How long can a Vidu Q4 video be?
Choose 3–16 seconds per generation. Use a shorter duration for a brief action, or allow more time for character entrances, interactions and shot changes.
How do I make a character speak?
Enable audio generation and describe the speaker, dialogue and emotion in your prompt. In reference-to-video mode, you can also add voice references to guide the character’s voice. Results depend on your inputs and descriptions.
Where do these example videos come from?
The videos are official examples from the Vidu Q4 website. They illustrate creative possibilities and do not guarantee the same results for every generation.
Where can I see my video after submitting?
After signing in and submitting, you will enter the Zopia creative canvas. View task progress and results there, then continue refining and creating.
What happens next for your characters?
Bring your reference images, describe the action and dialogue, and start shaping your first Vidu Q4 video.
Explore other AI video and image models on Zopia
Try the same brief with different models and keep the version that fits your channel.
Seedance 2.0 Pro
AI Video Generator
Free / Unlimited / Cinematic Quality
ByteDanceExplore model →
Nano Banana 2
AI Image Generator
Up to 10 references / consistent characters & products
GoogleExplore model →
Kling 4.0
AI Video Generator
Early access on Zopia · Turn imagination into a complete scene.
KuaishouExplore model →