genmoai/mochi-1

Official View on Replicate →

Mochi 1 preview is an open video generation model with high-fidelity motion and strong prompt adherence in preliminary evaluation

Capabilities

Seed

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`fps`	integer	Frames per second	`30`	min: 10, max: 60
`guidance_scale`	number	The guidance scale for the model	`6`	min: 0.1, max: 10
`num_frames`	integer	Number of frames to generate	`163`	min: 30, max: 170
`num_inference_steps`	integer	Number of inference steps	`64`	min: 10, max: 200
`prompt`	string	Focus on a single, central subject. Structure the prompt from coarse to fine details. Start with 'a close shot' or 'a medium shot' if applicable. Append 'high resolution 4k' to reduce warping	`"Close-up of a chameleon's eye, with its scaly skin changing color. Ultra high resolution 4k."`	—
`seed`	integer	Random seed	`—`	—

fps integer

Frames per second

Default: 30 min: 10, max: 60

guidance_scale number

The guidance scale for the model

Default: 6 min: 0.1, max: 10

num_frames integer

Number of frames to generate

Default: 163 min: 30, max: 170

num_inference_steps integer

Number of inference steps

Default: 64 min: 10, max: 200

prompt string

Focus on a single, central subject. Structure the prompt from coarse to fine details. Start with 'a close shot' or 'a medium shot' if applicable. Append 'high resolution 4k' to reduce warping

Default: "Close-up of a chameleon's eye, with its scaly skin changing color. Ultra high resolution 4k."

seed integer

Random seed

Version: 1944af04d098 Updated: 2/26/2026 3.2K runs