AI image to video

To turn an image into a video with AI, upload the photo, choose how it should move, add a few words about what happens if you like, check the box and press Animate the photo. A five-second clip at about 720p, with the natural sound of the place, comes back in a few minutes. Make it with any DreamBench plan.

Generating a video from an image takes a few minutes for one five-second clip. The page shows the clip as soon as it is ready. If the first model declines your photo, a second one tries once, which adds a little time, and if that one declines too, the credits for the run come back automatically.

No. Every run needs a DreamBench plan; you create an account on the way, and the button shows the price in credits before you press it. Canva and Google Photos make a limited number of clips without paying, silent or short; here every clip carries the sound of the place.

About 720p, in your photo's own shape: a vertical photo gives a vertical clip and a landscape photo a landscape one, so 9:16 and 16:9 both work. There is no 1080p or 4K option: the clip is 720p however large the photo, which is scaled to 2048 pixels on its long edge first. The result is an MP4 only and doesn't loop by itself.

No visible one: nothing is drawn over the picture. The MP4 carries a label in its file data saying it was made with AI on DreamBench, with the picture and sound untouched, so apps that read such labels can tell. Re-saving the clip in another app can drop that label.

Yes, and without paying. Canva's Image to Video makes a silent five-second clip within its monthly AI allowance, though Canva says faces may not move as expected. On an iPhone, the Google Photos app's Photo to video makes a limited number of clips a day, in some regions. This page runs in a phone's browser too, with no app to install.

No. The Ken Burns effect pans and zooms across a still photo, as iMovie on an iPhone does to the photos you add, so nothing in the picture itself moves. Here a video model moves the scene: hair in the wind, a blink, a turn of the head, water running, with the camera still or drifting.

Image to video starts from your picture: the first frame is your photo, so the people, the place and the framing are yours, and the model adds the motion. Text to video invents everything from a description. This page is image to video; the DreamBench composer makes whole films from a sentence, with your photos as references if you like.

Yes, it has sound: the run asks for the quiet natural sound of the place, with no voices and no music. The player starts muted; its sound button turns it on. Music isn't made here: lay a track over the MP4 in CapCut or iMovie, or carry the clip into the composer. If the second model makes the clip after a refusal, it is silent.

The run tries a second model once. If that declines too, your credits for the run come back automatically. A photo of a person never goes to Seedance, Veo or Grok Imagine. The models used here may still decline a photo of a child or of a well-known person.

Not on this page: each run takes one photo and makes one five-second clip. Under the result, Make it longer and Continue as a story open the DreamBench composer with your photo as a reference; more photos can be added there, and it joins several shots into one longer film with sound.

Because the model invents every frame after the first: a face can drift over five seconds, most when the head turns far or the motion is fast, and fingers can merge or multiply as they move. Start from a sharp photo, choose a gentle motion, keep the note to one action and compare the clip with your photo before you share it.

No. Nobody speaks, sings or mouths words in the clip: the run asks for that, and for only the quiet sound of the place. There is no voice or lip-sync setting, because making a real person appear to say something is not what this tool is for.

Your own, or photos you have permission to use, and never a photo of someone without their consent. You confirm it with a box before each run. Every clip is labeled as made with AI, and the link under it reaches us if a clip should be removed.

Each run uses 69 credits from your DreamBench plan. Starter, $9.99 a month, includes 1,100 credits a month, about 15 runs of the AI image to video, plus AI video. A run that fails is refunded.

Yes: a free account and a DreamBench plan. The tools are part of every plan, from Starter at $9.99 a month.

A photo up to 40 megabytes: JPEG, PNG, WebP, HEIC or another format your browser opens. A photo straight from a phone is fine, an iPhone's HEIC included: the page scales a large picture down to 2048 pixels on its long edge before the run, so the result is never larger than the page says.

Yes. Under the DreamBench Terms you may use what a tool makes for personal and commercial purposes, including advertising, monetized content and work for clients, with no separate commercial-use upgrade. The Terms don't guarantee copyright in a result or clear another person's rights in your upload, so only use photos you have the rights to.

No. DreamBench doesn't use your prompts, uploads or videos to train or fine-tune AI models, and any future training program would be opt-in.

It stays in your browser until you make a result. Then it is used to make that result and nothing else, and removed from the tool's storage after seven days along with the result. A result you keep in your library stays as long as you keep it.