Image to video AI

Start from a photo, describe how it should move, and ChatCut generates the clip. Then cut it, caption it and add sound on the same timeline.

Trusted by 400,000+ creators

How it works

How to turn an image into a video with AI

  1. 01

    Describe the motion

    Write what should move and what the camera does, then open ChatCut and sign in free.

  2. 02

    Add your image

    Switch to Video Gen, pick a model, and drop the photo into the first-frame slot. Set the aspect ratio, resolution and length.

  3. 03

    Generate and edit

    Check the credit estimate, send, and follow progress in My Assets. Drag the finished clip onto the timeline to keep editing.

Why ChatCut

Why turn images into video in ChatCut

  • Seedance and Kling in one menu

    Seedance 2.5, Seedance 2.0, Kling 4.0, Kling 3.0 and more sit in one Video Gen menu. Each model shows its own inputs, lengths and resolutions.

  • Control the first and last frame

    Use your photo as the first frame, add an ending frame, or attach reference images for a character, a set or a product.

  • The clip lands in your project

    Finished videos appear in My Assets. Download them, or drag them onto the timeline next to your footage.

  • Edit without switching apps

    Trim the clip, add captions, music or a voiceover, then export the finished video.

  • See the credit estimate first

    The estimate shows next to Send before anything runs. Each take is its own job and charge, so you decide when to try again.

  • Or ask the AI editor

    Describe the shot in plain words and the ChatCut Agent sets up the generation for you. Agent messages use credits too.

First and last frames guide this Kling 4.0 clip.

Tips

Plan motion for image to video AI

Start with a clear image, then describe one main movement and what should stay fixed.

  1. Prepare the starting frame

    Use a sharp image with one clear subject and crop to the final aspect ratio because the clip starts from that frame.

  2. Describe one main movement

    Specify who moves, direction, speed, and camera motion so the prompt focuses on changes the image does not already show.

  3. Add the intended final frame

    Use a supported last frame when the ending matters, keeping framing and light similar to reduce what the model must invent.

Use cases

What people make with image to video AI

  • Bring a still photo to life

    Give a portrait or a travel shot gentle motion: hair in the wind, a slow push-in, light sliding across the frame. This still came from GPT Image 2.6 and was animated with Kling 4.0.

  • Product shots from one photo

    Attach the product photo as a reference and name what must not change, such as the label. This coffee bag kept its wording through a Seedance 2.5 take.

  • Characters that stay the same

    Add character sheets and a set as reference images, name each one in the prompt, and reuse them for the next shot of the story.

  • Test a shot before you plan it

    Generate a key frame, animate it, and see whether the move works before you book a shoot or brief an animator.

Keep editing

One clip is the start of a video

Generate several shots, cut them together, add captions and music, and export the finished video from the same editor. Manual editing and export don't use credits. Need a clip from words alone? Try text to video.

New on YouTube Seedream 5.5 Is Now in ChatCut Seedream 5.5 in ChatCut
More from our channel All videos

Skills

Hundreds of skills for any video

100+ Skills. No experience needed. Start in one click.

Talking-Head Edit RecreationEdit your talking-head footage with the cuts and styling of a reference video. Share a talking-head video you like (upload it or paste a link), then add your own recording. AI breaks down the reference's pacing, captions and on-screen text, motion graphics, B-roll, framing and sound, and applies them to what you say. You get a finished video and an editable project. It uses your own footage throughout, with no face swaps or generated shots.Explainer VideosFor children’s stories, nature explainers and playful educational content. Enter a topic or full script to create a narrated animated short with clay characters and a handmade stop-motion look.ClaymotionProduct ads & UGCFor brand awareness and product launches. Provide product photos, brand details and display copy to create a city ad featuring your product on a Times Square billboard.Product in Times SquareViral EffectsUpload one athlete reference, and optionally a rooftop scene image, to create a continuous parkour shot that connects running, obstacle vaults and landings with a close-following camera.Urban ParkourProduct TV CommercialsFor highlighting candy and snack textures. Provide product images and ingredient details to create a commercial combining sugar-crystal particles, macro details and a final product shot.Sugar Crystal StormApp PromoCreate your own product promo with the AI Assistant Feature Tour template. Provide your website or screenshots, logo and key message. Get a tailored video with an editable timeline.AI Assistant Feature Tour

FAQ

Image to video AI FAQ

How does image to video work in ChatCut?

Switch the composer to Video Gen, pick a model, add your image as the first frame (and a last frame if you want one), describe the motion and send. The clip appears in My Assets for you to download or edit.

What happens when I click Animate it in ChatCut?

ChatCut opens in a new tab with your request. Sign in with Google or email within 30 minutes and the agent starts a new project from it. If you picked an example, the agent uses that image as the first frame, so there is nothing to upload; with your own photo, add it when the agent asks. The request asks the agent to show the credit estimate before it generates.

Which AI models can I use?

The Video Gen menu lists Seedance 2.5, Seedance 2.0, Seedance 2.0 Fast and mini, Gemini Omni, MiniMax H3 and H3 Max, Kling 3.0 and Kling 4.0. Inputs, lengths and resolutions differ by model; the Model Capability Matrix lists them.

Is image to video free?

No. Video generation is a Pro feature and uses credits, and ChatCut shows the estimate before you send. Trimming, captioning and exporting the result on the timeline don't use credits. Rates are in the credits policy.

Can I control how the video ends?

Yes. Models that support frames take a first and a last frame, so the clip moves from one image to the other. Keep both frames similar in framing and lighting for a smoother move.

What should I write in the prompt?

Say what should move, how the camera behaves and the mood, for example a slow push-in, hair moving in the wind, warm evening light. Then name what must stay the same, like the face or the product label.

How long can an image to video clip be?

It depends on the model. Kling 3.0 makes 3 to 15 seconds and Kling 4.0 up to 30 seconds in one clip. For anything longer, generate several shots and cut them together on the timeline.

Can I get more than one version?

Each successful generation gives you one video. Send the same prompt again for another take; every take is a separate job, asset and charge.

Can I edit the generated video?

Yes. Trim it, add captions, music or a voiceover, combine it with other clips and export, all in ChatCut. You can also turn photos into a slideshow without generating anything.