22 models, one interface
Switch model without rebuilding your settings. Anything the new model cannot honour is struck through with the reason rather than silently dropped.
- Ochra Recast
- Seedance 2.5
- Seedance 2.0
- Kling 3.0 Omni
Turn a photo into a video clip. Describe the motion, choose the model, length and resolution, and see the credit price before you generate.
Bring a still image, describe the motion, and get a video back. The picker spans several model families with different strengths — among them Seedance for long takes with sound, WAN for prompt fidelity and first-frame animation, and Grok for always-on native audio — and each model shows its own credit price for the duration and resolution you have selected before you generate.
Switch model without rebuilding your settings. Anything the new model cannot honour is struck through with the reason rather than silently dropped.
Duration is a per-model capability, not a global setting. The picker shows which lengths each model actually supports.
Choose the output resolution where the model allows it. Some are locked to a single resolution by the provider; that is shown on the row.
Dialogue and effects generated with the picture rather than dubbed over it, on the models that support it.
Pin the opening frame so a character you already generated stays the same character. Several models also accept a closing frame.
The credit cost updates as you change model, duration and resolution — and the figure on screen is the figure charged.
Upload the still you want to animate, or start from a template.
Describe the motion, not the frame — what moves, how the camera behaves, how long it lasts.
Pick a model, check the credit cost, and generate. The result lands in your gallery.
By constraint. A long take with sound: Seedance 2.5. A faithful animation of a first frame you already have: WAN. Native audio on every clip: Grok Imagine Video. Every model row states what it is good at and what it costs.
Up to 30 seconds with Seedance 2.5, Wan 3.0 and Wan 3.0 Prime. Other models cap lower — the picker shows each model’s supported durations and strikes through the ones it cannot produce.
On the models that support it, yes — audio is generated together with the video. Models without audio support are marked in the picker.
From 3 credits (≈ $0.17) for the shortest, lowest-resolution clip on the least expensive model. It depends on the model, the resolution and the duration, and the price is shown under the Generate button before you commit.
Yes. As between you and Ochra, you own what you generate and you may use it commercially — for marketing, social media, advertising and business use. Where a model provider's own terms are stricter than ours you must follow them too, and AI output may not be protected by copyright in every country.