For a selected performance: the site’s video templates
Choose jeanphilai.app when the template preview already shows the movement you want. Sign in with a verified account, select a scene and ensure your balance covers the quote, buying credits if needed. Select a permitted photo, confirm your rights, then click Upload. Wait for both the photo and template reference checks to pass and approve the photo before generation. The photo is used directly without a separate generated-portrait charge. The featured 19-second templates cost $9.90, or 990 credits, and deliver silent HD 720p video. Watch the actual template previews. Compare credit packs.
Check each card’s duration and the generation quote. Paris Fashion Week is about 17.8 seconds and costs $9.90; its name does not make it a newly choreographed runway walk. Longer scenes have different quotes. Use the preview to decide whether you want that specific performance, rather than choosing only by the scene’s title.
For movement you describe: image-to-video
An image-to-video model uses a portrait or scene as its visual starting point and a prompt to describe the action. Runway’s Gen-4 documentation requires an input image and recommends focusing the text on motion. The Hailuo image-to-video workflow likewise starts with an image and a movement description. Kling’s documented image-to-video endpoints accept that same basic input pair, although their settings differ. Runway: creating with Gen-4 Video. Hailuo: image-to-video workflow. Kling 2.6 Pro on Higgsfield: documented inputs.
Use this route for an original small jab, head turn or walk whose exact timing can be invented. Choose a model that supports your intended duration and format, then test one uncomplicated action. A prompt requesting a dance does not reproduce a particular recording’s choreography or camera movement.
For a recorded move: image plus motion reference
Higgsfield’s Kling 2.6 Motion Control documentation describes animating a source image from a provided video. This is a relevant input format when a specific dance, gesture sequence or camera performance matters. It is a documented capability, not a guarantee that the face, background and every frame will exactly match your intention. Higgsfield: Kling 2.6 Motion Control.
On this site, Create your own → Use my own motion clip accepts a licensed 3–30 second reference. A new portrait costs $1.00. Motion generation costs $9.90 for up to 19 seconds; a longer reference is charged by its rounded-up whole-second duration, with the final cent amount rounded up. For example, 20 seconds costs $10.43 for motion and 30 seconds costs $15.64, before a new portrait charge.
For the starting portrait: image generation or editing
Midjourney supports image inputs and different reference modes. Google’s Nano Banana image models support text-and-image creation and editing. Use a permitted reference when you want to carry a character’s appearance into a new setting, and inspect the new image before animating it. Midjourney’s Style Reference controls visual style; its documentation explicitly distinguishes that from copying objects or people. Midjourney: web creation and image inputs. Midjourney: what Style Reference controls. Google: Nano Banana image generation.
Save one accepted portrait as your visual reference rather than independently inventing the face for every scene. Repeating a hair and suit description can help communicate the intended look, but wording alone is not an identity lock. Check reference support for the exact model version you select.
Compare the total work, not just a headline price
Count the image stage, video stage and any extra iterations separately. Also check the clip length, resolution, watermark, export access and permitted inputs. A low price for a short trial clip does not directly compare with a longer template video, and a higher-resolution model does not automatically preserve a difficult gesture more accurately.
Free trials can help explore a workflow, but their available models and allowances change. Choose the tool you can afford to use for your intended output, read its current terms and confirm the quote before rendering. Prepare a clear portrait and a modest action before spending credits on more variations.
A practical selection for three common projects
You want your photo in Apartment shadowboxing: watch that 19-second template and use its $9.90 direct-photo route. You want an original character slowly bowing in a café: prepare a portrait plus a small motion prompt, then choose an image-to-video renderer. You want an original character performing your recorded dance: prepare a suitable portrait and a licensed reference clip for a motion-control workflow.
For this site’s idea route, the new portrait is $1.00 and the five-second silent HD video is $18.90 after you approve the portrait. You can instead copy the prepared portrait and motion prompts without signup and take them to a tool you already use. Choose the input path first; it determines what control you have over the resulting scene. Follow the complete photo and idea workflows.