You're looking at a specific version of this model. Jump to the model overview.

microsoft /vq-diffusion:81ca56aa

Input

string

Choose generating images from in-the-wild text, MSCOCO datasets, or ImageNet class label.

Default: "in-the-wild text"

string
Shift + Return to add a new line

Prompt for generating image. Valid when generation_type is set to in-the-wild text and MSCOCO datasets.

Default: ""

string

Choose the ImageNet label. Valid when generation_type is set to ImageNet class label.

Default: "None"

integer
(minimum: 1, maximum: 8)

Number of generated images.

Default: 4

number
(minimum: 0, maximum: 1)

Sample with truncation.

Default: 0.86

number

Improved VQ-Diffusion with learnable classifier-free sampling.

Default: 1

Output

image
image
image
image
Generated in