Create
AI Video Generator: from a prompt to a finished clip
Describe a shot, start from an image, or edit a clip you already have. Pick a model, set the length, resolution and shape, and Agent One delivers the video in your chat.
AI Video Generator turns a written prompt, a still image or a clip you already have into a short video, with several video models to choose from in one panel. Set the length, resolution, shape and audio, guide it with reference media or first and last frames, or describe a change to an existing clip. It's built for musicians, creators and marketers who need release visuals, promos and social clips without a camera or an editing timeline.
See it in action
- How to Do the Viral Character Replacement Trend with AI | Soundverse AI x Seedance | AI Music Video A walkthrough of the viral character-replacement trend, where every person in a clip becomes the subject of one selfie, made with AI Video Generator.
- Every AI Tool in Soundverse "Create": Songs, Vocals, SFX, Videos and More | Full Walkthrough From 2:57. The AI Video Generator part of Soundverse's Create walkthrough: a short video from a prompt — Seedance, Veo or Gemini Omni.
How it works
Choose a model and describe it
Open AI Video Generator under Create in AI Tools, pick a model, and describe the scene, subject, motion and mood. Enhance can rewrite the prompt for you.
Add images or a clip
Optionally add reference images, video or audio from your Library, depending on the model, or turn on Use first / last frame to start and end on images you choose. To change a clip you already have, pick the edit model and say what to change.
Set the output
Choose the duration, resolution and aspect ratio the model offers, and whether to generate an audio track. The panel shows the estimated token cost before you generate.
Generate and download
Click Generate in Chat and Agent One renders the clip in the chat, where you can play it, go full screen and download it. It's also saved to Videos in your Library.
Why use AI Video Generator
Several models, one panel
Pick from several video models in the Model picker. The panel shows only the settings the chosen model supports, and it flags any combination the model can't render before the run starts.
Start from text, image or clip
Generate from a prompt alone, guide it with reference media, pin the first and last frames, or edit a video you already have.
Control length and shape
Make clips from 3 to 30 seconds depending on the model, at up to 1080p, in widescreen, vertical and, on some models, square and other shapes.
Sound when you want it
Turn on Generate audio to get a soundtrack with the clip on models that offer it. Some models always include audio.
Great for
- Social clips, Reels and Shorts
- Promo and teaser videos for a release
- Visual loops and backdrops for a track
- Animating cover art into motion
- Restyling footage you already have
- Trend videos like character replacement
Frequently asked questions
What is AI Video Generator?
AI Video Generator is a Soundverse tool that turns a text prompt, a still image or reference media into a short video clip, with several video models to choose from in one panel. It can also edit a clip you already have from a description of the change, and it runs through Agent One, so the finished video appears in your chat.
How long can AI-generated videos be?
Each AI Video Generator run makes one clip of 3 to 30 seconds, depending on the model. The default model makes 4 to 15 seconds; other models offer 3 to 10 seconds, 4 to 30 seconds, or fixed 4, 6 or 8 second clips. For a longer piece, Multi-Scene Video Builder joins up to 10 scenes into one video.
Can I turn an image into a video?
Yes. Turn on Use first / last frame in AI Video Generator and pick an image from your Library as the first frame, and the video starts from it. Add a second image as the last frame to set where the shot ends. First-frame mode works with every model that generates new clips. You need the rights to any image you use.
Can AI Video Generator edit a video I already have?
Yes. Pick the edit model, choose an MP4 from your Library as the Video to edit, and describe the change in the What to change box, such as making it snow or turning a jacket red. The edit keeps the original clip's length and aspect ratio. Source clips can be up to 30 seconds and 200 MB.
What resolutions and aspect ratios does AI Video Generator support?
AI Video Generator renders at 480p, 720p or 1080p depending on the model, and it doesn't generate 4K; not every model offers 1080p. The default model supports 16:9, 9:16, 1:1, 4:3, 3:4 and 21:9, while some models film only widescreen 16:9 or vertical 9:16. The panel shows each model's own options.
Is AI Video Generator free?
No. AI Video Generator is available on the Creator plan and above, and each run uses tokens from your plan. The cost depends on the model, the clip length, the resolution and whether audio is generated, and the panel shows the estimate before you generate. You can sign up for Soundverse free and explore the core tools before you upgrade.
Can I use AI-generated videos commercially?
AI Video Generator runs on paid plans, and paid plans add a commercial license; the Pricing page lists what each plan covers. You need the rights to anything you upload or pick as input, including reference images, video and audio, first and last frames, and any clip you edit. Only use a face or voice you own or have permission to use.
Can I add music to an AI video?
Yes, in a few ways. Some models generate an audio track with the clip. After a run, the Add a soundtrack suggestion asks Agent One to score it. To make visuals for a finished song instead, AI Music Video Generator writes a script for your track, films the scenes and cuts them to the music.
Ready to make your next clip?
Describe the shot, add an image or a clip, and let Agent One render it at up to 1080p.