nateraw/stable-diffusion-videos

Generate videos by interpolating the latent space of Stable Diffusion

Capabilities

No capability data available

Community model (estimated from hardware time)

Name	Type	Description	Default	Constraints
`fps`	integer	Frame rate for the video.	`15`	min: 5, max: 60
`guidance_scale`	number	Scale for classifier-free guidance	`7.5`	min: 1, max: 20
`num_inference_steps`	integer	Number of denoising steps for each image generated from the prompt	`50`	min: 1, max: 500
`num_steps`	integer	Steps for generating the interpolation video. Recommended to set to 3 or 5 for testing, then up it to 60-200 for better results.	`50`	—
`prompts`	string	Input prompts, separate each prompt with '\|'.	`"a cat \| a dog \| a horse"`	—
`scheduler`	string	Choose the scheduler	`"klms"`	default ddim klms
`seeds`	string	Random seed, separated with '\|' to use different seeds for each of the prompt provided above. Leave blank to randomize the seed.	`—`	—

fps integer

Frame rate for the video.

Default: 15 min: 5, max: 60

guidance_scale number

Scale for classifier-free guidance

Default: 7.5 min: 1, max: 20

num_inference_steps integer

Number of denoising steps for each image generated from the prompt

Default: 50 min: 1, max: 500

num_steps integer

Steps for generating the interpolation video. Recommended to set to 3 or 5 for testing, then up it to 60-200 for better results.

Default: 50

prompts string

Input prompts, separate each prompt with '|'.

Default: "a cat | a dog | a horse"

scheduler string

Choose the scheduler

Default: "klms"

default ddim klms

seeds string

Random seed, separated with '|' to use different seeds for each of the prompt provided above. Leave blank to randomize the seed.

Version: 2d87f0f8bc28 Updated: 2/26/2026 58.5K runs