You're looking at a specific version of this model. Jump to the model overview.
zf-kbot /image-upscaler-20251103:e3e84158
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 |
|---|---|---|---|
| input_image |
string
|
输入图片文件(支持jpg、png、bmp格式)
|
|
| max_long_side |
integer
|
1920
Max: 4096 |
目标最长边尺寸(推荐1280-1920)。图片会等比缩放到此长边,短边padding到128倍数,处理后自动裁切黑边。设为0表示不调整
|
| aspect_ratio_threshold |
number
|
1.5
Min: 1 Max: 4 |
长宽比阈值(主动OOM预防)。当图片长宽比<此值时,自动降低max_long_side以节省显存。推荐1.5-2.0
|
| downgrade_max_long_side |
integer
|
1280
Min: 128 Max: 2560 |
降级后的max_long_side。当图片长宽比<阈值时使用此值。推荐1280
|
| face_enhance |
integer
|
0
Max: 1 |
启用GFPGAN面部增强。0=禁用,1=启用。可改善人脸细节
|
| sd_post_process |
integer
|
1
Max: 1 |
启用SeedVR2高级后处理。0=禁用,1=启用。可进一步提升画质
|
| sd_resolution |
None
|
2k
|
SeedVR2目标分辨率预设。2k=2560×1440,4k=3840×2160。基于此计算目标最短边
|
| sd_min_short_side |
integer
|
0
Max: 4096 |
SeedVR2最短边上限(0=仅使用预设)。设置后会取预设计算值和此上限的较小值,用于限制显存占用
|
| post_upscale |
integer
|
0
Max: 1 |
启用4x超分辨率模型(ClearRealityV1_Soft)。0=禁用,1=启用
|
| scale |
number
|
1
Min: 1 Max: 4 |
额外放大倍率(在max_long_side调整后)。通常保持1.0。设置>1.0会在max_long_side基础上再放大
|
| image_output_format |
None
|
jpg
|
输出图片格式(jpg压缩率高,png无损)
|
| log_level |
None
|
info
|
日志详细程度。info=基本信息,debug=详细调试信息
|
Output schema
The shape of the response you’ll get when you run this model with an API.
{'format': 'uri', 'title': 'Output', 'type': 'string'}