Reference-to-video lets you combine up to 15 images to lock in characters, wardrobe, props, products and locations, plus up to 3 short mp3 voice clips to keep a character's voice consistent. Refer to each picture by its tag in the prompt. This clip was made with Vidu Q4 Preview on GoEnhance from two references: a mug and a canyon scene.
ShengShu says Q4 coordinates facial expression, emotion, body movement and voice, so characters act rather than just move. Dialogue and sound effects are generated together with the picture, and you can switch audio off at the same price. Put spoken lines in quotes in your prompt.
Vidu describes steadier camera moves and cuts that follow your direction, with the camera staying on the action in fights and chases and effects such as explosions blending in more naturally. Describe the camera move in the prompt, for example a tracking shot or a slow push-in.
Choose 540p, 720p, 1080p, 2K or 4K, with 10-bit color at the top end, and any length from 3 to 16 seconds. In image-to-video the clip keeps your picture's aspect ratio, so wide shots stay wide. This train clip was made with Vidu Q4 Preview on GoEnhance at 540p from a 21:9 still.
Vidu Q4 Preview animates a still, so start with a clear image in the shape you want; the video keeps its aspect ratio. To build a scene from several pictures, open reference to video instead.
Describe what happens, how the camera moves and what we hear. Put dialogue in quotes and keep each line short so the delivery stays natural.
Choose 3 to 16 seconds and 540p to 4K. Each second costs 6 tokens at 540p, 12 at 720p, 16 at 1080p, 25 at 2K and 51 at 4K, so a 5-second 720p clip is 60 tokens.
Draft at 540p, which works on the free plan, then re-render the version you like at a higher resolution with your membership.
Bring a still to life with dialogue, sound and camera moves, or build a scene from up to 15 references. Start free at 540p and scale up to 4K when you're ready.
Try Vidu Q4 Preview