Framepack Image-To-Video | Autoregressive Video Generation

Dashboard Explore Inspiration

Home/Explore/wavespeed-ai/framepack

image-to-video

wavespeed-ai/framepack

Framepack is an efficient autoregressive Image-to-Video model that generates smooth, temporally consistent videos from a single image. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Documentation

image*

Hint: You can drag and drop a file or click to upload

prompt*

Animate this scene with a deliberate and unfolding sense of drama. The camera should begin with a tight close-up on the woman's intensely focused eyes, before slowly and dramatically pulling back to reveal her full figure standing on the rain-slicked balcony. As the camera pulls back, a sudden gust of wind should whip through the scene, causing her crimson gown to billow dramatically around her, and strands of her dark hair to fly across her face. She should lift one hand slowly and deliberately, as if anticipating or reacting to something in the cityscape below. The neon lights of the metropolis in the background should flicker erratically, casting dynamic and shifting shadows across the balcony and her form. A distant siren should wail, its sound seemingly carried on the wind. The rain on the balcony floor should briefly intensify, creating visible ripples and reflections of the dramatic lighting. The overall animation should build a sense of heightened tension and impending conflict through its deliberate and impactful movements.

negative_prompt

aspect_ratio

resolution

seed

num_inference_steps

num_frames

guidance_scale

Enable Safety Checker

Idle

Your request will cost $0.198 per run.

For $10 you can run this model approximately 50 times.

One more thing::

ExamplesView all

README

Framepack

FramePack is a next-frame (next-frame-section) prediction neural network structure that generates videos progressively.

FramePack compresses input contexts to a constant length so that the generation workload is invariant to video length.

FramePack can process a very large number of frames with 13B models even on laptop GPUs.

FramePack can be trained with a much larger batch size, similar to the batch size for image diffusion training.