Image-to-video AI takes a single still photo and generates a short video from it — the camera moves, the water ripples, the subject turns their head. You provide the starting frame; the model invents the motion.
It's one of the most practical AI video workflows because you already control the most important variable: the composition. Instead of gambling on what a text prompt produces, you start from an image you've approved and only ask the AI to animate it.
Who it's for
- Product marketers — turn a clean product shot into a rotating showcase clip for ads or landing pages.
- Family archivists — bring old photos to life with subtle, natural motion.
- Social media creators — convert existing photos into scroll-stopping video posts without a camera crew.
Step-by-step: photo to video on Bno AI
- Open the AI Video Generator and switch to image-to-video mode.
- Upload your photo. JPG, PNG, and HEIC are supported. Use a clear, high-resolution image with a well-defined subject — the model can only animate what it can see.
- Describe the motion. Write a short prompt about what should move, not what's in the picture. More on this below.
- Pick a model. Seven video models are integrated; the next section helps you choose. The credit cost for your exact settings is shown before you confirm.
- Generate. Generation usually takes a few minutes depending on the model and settings. A progress indicator keeps you posted.
Writing motion prompts: describe the "how," not the "what"
The model already sees your image — it knows there's a gondolier, a sneaker, or a grandmother in the frame. Repeating that wastes your prompt. Spend it on motion instead:
- ❌ "A red sneaker on a white background" (the model knows)
- ✅ "Rotate the shoe slowly, keep everything else still"
- ✅ "The camera moves slowly forward, golden light shifts across the buildings"
- ✅ "She smiles gently and turns her head toward the window"
Three rules of thumb: name one primary motion, specify the camera behavior (static, slow push-in, pan left), and say what should stay still. Constraints prevent the model from animating everything at once.
Which video model should you pick?
Bno AI integrates seven video models. A one-line orientation for each:
| Model | Best for |
|---|---|
| Hailuo 2.3 | Strong, fluid motion that follows camera directives — a solid default |
| Kling v3 | Clips that need sound — the one model with an audio generation option |
| Wan 2.7 | Dependable short clips at a value price |
| Seedance 2.0 | Steering the result with a reference image, up to 1080p |
| Sora 2 | Physically plausible motion and longer takes (up to 12s) |
| Grok Imagine 1.5 | Testing a motion idea cheaply — clips start at 9 credits |
| Veo 3.1 Fast | Cinematic polish on a fixed 8-second clip |
Don't overthink the first pick. Generate with one model, and if the motion style isn't right, the same image and prompt are easy to re-run elsewhere.
Troubleshooting common problems
The subject's face or body distorts. This usually means the motion you asked for is too aggressive for the source image. Use a higher-resolution photo with a clearly defined subject, and dial the prompt down to one subtle motion.
Everything moves too much. Add explicit constraints: "keep the background still," "only the hair moves in the wind." Models fill silence with motion — if you don't say what stays put, nothing will.
The clip feels too short. Clip length depends on the model and the duration you select. Longer durations cost more credits — the exact cost for your settings is always displayed before you generate, so you can compare before committing.
A note on credits
Video is charged in credits, not locked behind a subscription — any account with enough balance can generate. The Free plan's 10 daily credits cover two Sora 2 generations, one budget Grok Imagine clip (6 seconds at 480p, 9 credits), or about five GPT Image 2 images for testing source frames. Video prices currently range from 5 credits for Sora 2 to 269 credits for Kling v3 at 1080p, 10 seconds, with audio — so for regular video work, Pro ($20/month, 2,000 credits) or Ultimate ($40/month, 5,000 credits) is where the volume comes from.
FAQ
How long does it take to generate a video? Generation usually takes a few minutes, depending on the model, duration, and resolution you choose.
Can I use the generated videos commercially? Commercial-use rights depend on applicable law, your input rights, the selected model or provider, and Bno AI's Terms. Check those rules before using an output in marketing or another business context.
Does it work with any photo? It supports JPG, PNG, and HEIC. For the best results, use clear, high-resolution images where the subject is well-defined.
Can I try image-to-video on the free plan? Yes. The free tier's 10 daily credits cover one Grok Imagine 1.5 clip (6 seconds at 480p, 9 credits) — enough to animate one photo a day at no cost. Heavier models need a larger credit balance; open the video generator to see each model's exact price for your settings.