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

nsfw-api /hunyuan-custom:d63829f0

Input

string
Shift + Return to add a new line

Positive prompt (comma‑separated)

Default: "Realistic, High‑quality. A woman is drinking coffee at a café."

string
Shift + Return to add a new line

Negative prompt

Default: "Aerial view, overexposed, low quality, deformation, bad hands, distortion, text, black border."

string
Shift + Return to add a new line

Video resolution as <height>x<width>

Default: "512x896"

integer
(minimum: 2)

Number of frames to sample (default 129)

Default: 129

integer
(minimum: 1, maximum: 50)

Denoising steps

Default: 30

integer

Random seed

Default: 1024

boolean

Use FP8 mixed‑precision weights (reduces VRAM)

Default: false

boolean

Off‑load parts of the model to CPU (helps <80 GB GPUs)

Default: false

file

Foreground‑segmented PNG to condition generation. If omitted, the default demo image from the repo is used.

Output

No output yet! Press "Submit" to start a prediction.