zf-kbot/seedvr2

Public
5.9M runs

Run zf-kbot/seedvr2 with an API

Use one of our client libraries to get started quickly. Clicking on a library will take you to the Playground tab where you can tweak different inputs, see the results, and copy the corresponding code to use in your own project.

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
apikey
string
API Key
input_image
string
输入图像
target_megapixels
number
3.6

Min: 2

Max: 100

目标像素(百万像素)
ttp_switch
boolean
False
是否分片
tiles_per_row
integer
3

Min: 3

Max: 10

每行分片数量
short_pixels_after_tile
integer
1024

Min: 512

Max: 4096

分片压缩后短边
upscale_multiple_after_tile
number
2

Min: 1

Max: 4

分片压缩放大比例
website_name
string
ImgUpscaler.AI
Website name for filename suffix
image_filename
string
test
Custom filename for the output (without extension)
output_ext
None
.jpg
输出图像格式
quality
integer
90

Min: 1

Max: 100

输出图像质量 (1-100)
encode_tiled
boolean
False
VAE 编码是否分片
encode_tile_size
integer
1536

Min: 512

Max: 2048

VAE 编码分片大小
encode_tile_overlap
integer
128

Max: 256

VAE 编码分片重叠
decode_tiled
boolean
True
VAE 解码是否分片
decode_tile_size
integer
1536

Min: 512

Max: 2048

VAE 解码分片大小
decode_tile_overlap
integer
128

Max: 256

VAE 解码分片重叠
swap_io_components
boolean
False
是否交换 IO 组件
blocks_to_swap
None
0
要交换的块数量,0/32
offload_device
None
cuda:0
模型 offload 设备 (none=不卸载, cpu=卸载到CPU启用缓存, cuda:0=卸载到GPU启用缓存)
cache_model
boolean
True
是否缓存模型 (需要 offload_device=cpu 才生效)
upscale_method
None
bicubic
缩放算法 (bicubic支持GPU加速,lanczos质量最好但仅CPU)
resize_device
None
gpu
图像缩放设备 (bicubic/bilinear支持gpu,lanczos仅cpu)

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema
{
  "type": "array",
  "items": {
    "type": "string",
    "format": "uri"
  },
  "title": "Output"
}