You're looking at a specific version of this model. Jump to the model overview.
Input schema
The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.
| Field | Type | Default value | Description |
|---|---|---|---|
| image |
string
|
Reference image. Alpha is composited over gray; preprocessed images can be passed unchanged.
|
|
| prompt |
string
|
A 360-degree turning and circling video of a 2D RPG game character. Pixel Art. Game Sprite. This is a full-body long shot with a plain gray background. Preserve character's reference proportions, clothes and hairstyle. Clear readable silhouette, clean contours, flat local colors with restrained cel shading. Fixed camera, constant character size, feet on the same spot. aesthetic score: 5.5. motion score: 3.0. There is no text in the video.
|
Describe the video. Default is Sprute character turntable.
|
| negative_prompt |
string
|
3D, maya, render, blender, photorealistic, glossy plastic, sculpted figurine, dramatic lighting, gradients, walking, running, dancing, camera movement, zoom, cropped feet, outfit change, extra limbs, text, scene change, dithering,
|
None
|
| width |
integer
|
192
Min: 32 Max: 1024 |
None
|
| height |
integer
|
256
Min: 32 Max: 1024 |
None
|
| num_frames |
integer
|
81
Min: 5 Max: 161 |
Must be 4n+1.
|
| fps |
integer
|
24
Min: 1 Max: 60 |
Playback FPS; does not change sampled motion.
|
| steps |
integer
|
8
Min: 2 Max: 100 |
None
|
| switch_step |
integer
|
3
Min: 1 Max: 99 |
Step where high-noise model hands off to low-noise model.
|
| guidance_scale |
number
|
1
Max: 20 |
None
|
| low_guidance_scale |
number
|
1
Max: 20 |
None
|
| shift |
number
|
5
Min: 0.01 Max: 20 |
None
|
| low_shift |
number
|
5
Min: 0.01 Max: 20 |
None
|
| sampler_name |
None
|
uni_pc
|
None
|
| scheduler |
None
|
simple
|
None
|
| prepared_input |
boolean
|
False
|
Preserve exact opaque RGB pixels. Image must match width/height.
|
| reverse_frames |
boolean
|
False
|
Reverse output ordering. Sprute currently reverses AniSora output before direction extraction.
|
| output_format |
None
|
mp4
|
Output file format: MP4 (H.264), WebM (VP9), or looping animated WebP. Does not change inference.
|
| output_quality |
integer
|
80
Min: 1 Max: 100 |
Compression quality: higher means better fidelity and usually larger files. Codec-relative, not a model quality score; 100 does not guarantee lossless output. Use PNG frames for lossless originals.
|
| return_frames |
boolean
|
True
|
Return lossless PNG ZIP for local background removal and sprite assembly.
|
| seed |
integer
|
None
|
Output schema
The shape of the response you’ll get when you run this model with an API.
{'properties': {'frames': {'format': 'uri',
'nullable': True,
'title': 'Frames',
'type': 'string'},
'metadata': {'format': 'uri',
'title': 'Metadata',
'type': 'string'},
'seed': {'title': 'Seed', 'type': 'integer'},
'video': {'format': 'uri', 'title': 'Video', 'type': 'string'}},
'required': ['video', 'metadata', 'seed'],
'title': 'Output',
'type': 'object'}