infinite-talk/image-to-video
Create photo-based, speech-aligned videos with natural motion
| Parameter | Required | Type | Default | Range / Options | Description |
|---|---|---|---|---|---|
| prompt* | Yes (*) | string | — | — | Motion and camera description for the still |
| image* | Yes (*) | string (image URI) | — | — | Input still to animate |
| duration | No | integer | 5 | 4–30 (seconds) | Clip length |
| generate_audio | No | boolean | true | true / false | Native synchronized audio |
Seedance 2.5 4K Image to Video bills on generated video length:
Create photo-based, speech-aligned videos with natural motion
Sync an existing face video to a new audio track for dubbing.
Transform still visuals into cinematic motion clips with smooth, realistic transitions and creative flexibility.
Generate high quality videos from text prompts using Luma Ray 2.
Create lifelike videos from voices with accurate sync and adaptive dubbing.
Wan 3.0 Text To Video makes cinematic clips from prompts with audio
Seedance 2.5 4K Image to Video is a ByteDance image-to-video endpoint that animates a single still into a 4K clip. It suits product hero shots, campaign cutdowns, and character beats where texture like hair, fabric, or brushed metal needs to stay readable at full size.
The input and prompt behaviour are the same across resolutions; Seedance 2.5 4K Image to Video locks the output at 4K instead of 480p, 720p, or 1080p. Use a cheaper tier to test framing and motion, then rerun the same still and prompt here for the delivery version.
Seedance 2.5 4K Image to Video takes a prompt describing motion and camera, an image URL for the still to animate, duration from 4 to 30 seconds with a default of 5, and generate_audio. There is no aspect_ratio field, because the output frame follows your input image.
Feed a sharp, well-exposed still at or above the target frame size. Soft, heavily compressed, or already-enlarged sources limit what Seedance 2.5 4K Image to Video can resolve, since the model animates the detail present in your input rather than inventing missing texture.
Seedance 2.5 4K Image to Video is billed at $1.79 per second of generated video, scaling linearly with the duration you choose. Generations draw down your RunComfy usd / credits balance.
Yes. Try it in the RunComfy model UI first, then call the same Seedance 2.5 4K Image to Video template over the RunComfy API with matching parameters and a publicly reachable image URL. Both routes bill against the same usd / credits balance.
RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.
