Sound made with the picture
Turn on sound and Veo 3.1 adds dialogue, effects, and ambience that match the scene, in the same MP4 file.
Pick Veo 3.1 in Manus, describe the shot, and get a 4, 6, or 8 second video with sound. Start from your own image, or set the first and last frame.
Why Veo 3.1
Turn on sound and Veo 3.1 adds dialogue, effects, and ambience that match the scene, in the same MP4 file.
Give Veo 3.1 one image and the video starts from it, or set a first and last frame to control where it ends.
Choose 4, 6, or 8 seconds in 720p, or 8 seconds in 1080p, in 16:9 or 9:16. Veo 3.1 Fast has the same options.
A closer look
Every video below was made with Veo 3.1 or Veo 3.1 Fast in Manus. Each card shows the exact prompt.
Write a line of dialogue in the prompt and Veo 3.1 voices it, with sound that fits the scene.
Prompt
Night, a tiny ramen stall in a narrow rainy Tokyo alley lit by paper lanterns. Medium close-up of an older ramen chef with a white headband behind a steaming counter. He ladles rich broth into a bowl, sets it in front of the camera, smiles, and says in a warm, gravelly voice: "Last bowl of the night. Eat it while it's hot." Steam rises through the warm lantern light, rain falls softly behind him. Cinematic, shallow depth of field, film look. Audio: his line of dialogue, rain pattering on the awning, the clink of the bowl on wood, a quiet sizzle, distant city hum, no music.
Upload one image and Veo 3.1 starts the video from it, keeping the scene and its colors.

Prompt
Bring this photo to life. The rain keeps falling and ripples spread across the puddles on the cobblestones, the woman with the red umbrella walks slowly past the café, the waiter steps out and sets a cup on a table under the awning, and the warm window light flickers softly. The camera makes a slow, gentle push-in toward the café. Keep the scene, colors and composition of the photo. Audio: steady rain on the awning, footsteps on wet stone, a cup and saucer clinking, and a faint accordion melody drifting from inside.
Set a first and last frame, and Veo 3.1 fills in the motion between them.


Prompt
Animate the transformation from the first frame to the last frame: watercolor paint blooms across the pencil sketch on its own, starting in the sky and spreading down over the mountains, the pines and the lake, colors soaking into the paper in soft wet washes until the finished painting appears. The paper and the camera stay perfectly still, top-down, like a time-lapse of an artist's page. Audio: soft brush strokes and water dripping on paper, a gentle piano melody.

Choose 9:16 for Reels, Shorts, and other vertical feeds. This clip was made with Veo 3.1 Fast in 720p.
Prompt
Vertical action clip, low-angle follow cam: a skateboarder in a yellow hoodie drops into a sunlit concrete skate bowl, carves along the curved wall, pops a frontside air at the lip against a clear blue sky, and lands smoothly while the camera whips to follow. Late afternoon sun, long shadows, lens flare, energetic social-media style. Audio: skateboard wheels rumbling on concrete, the clack of the tail on the coping, a short cheer from friends off-screen, rushing wind.
Specs
What you can set when you pick Veo 3.1 or Veo 3.1 Fast.
How it works
01
Open Video in Manus and pick Veo 3.1 from the model menu below the input box.
Jelajahi Video02
Write what should happen and choose 16:9 or 9:16. Add one image to start from, or a first and last frame.
03
Ask Manus for changes in the chat, or open the video in the Video Editor in Manus Desktop to adjust clips, captions, and tracks.
Jelajahi Video EditorChoose the right model
Every video model in Manus makes 16:9 or 9:16 MP4 videos. Here is when to pick each one.
ByteDance
Pick it for videos of up to 30 seconds, guided by up to 30 reference images.
ByteDance
Pick it for 1080p videos of up to 15 seconds. A Fast version is included.
Pick it to edit a video you already have. Veo 3.1 only makes new videos.
MiniMax
Pick it for 5 to 15 second videos that follow up to 9 images, 3 video clips, and 3 audio clips.
See everything Video can do, edit on a full timeline, or compare all models.
Veo 3.1 is a video model from Google. In Manus it makes 4, 6, or 8 second videos in 720p, or 8-second videos in 1080p, in 16:9 or 9:16, with optional sound. It can start from one image or move between a first and last frame.
Open Video in Manus, pick Veo 3.1 from the model menu below the input box, and describe the video you want. You can also start your request with "Use Veo 3.1".
Availability and credit usage depend on your Manus plan. See pricing
Veo 3.1 Fast is the Fast version of Veo 3.1, also from Google. In Manus it has the same options: 4, 6, or 8 seconds, 720p or 1080p, 16:9 or 9:16, optional sound, and one starting image or a first and last frame. Pick it from the same model menu.
No. Veo 3.1 is a newer version of Veo 3 from Google. In Manus, Veo videos are made with Veo 3.1 or Veo 3.1 Fast. Veo 3 is not in the model menu.
Veo 3.1 makes 4, 6, or 8 second clips in up to 1080p and can start from one image or move between a first and last frame. Seedance 2.5 makes 4 to 30 second videos in 720p and follows up to 30 images, 10 video clips, and 10 audio clips.
Yes. Choose 9:16 for vertical videos or 16:9 for widescreen. 1080p is available for 8-second videos, and 4 or 6 second videos are 720p. Square videos and 4K are not available.