logo
RunComfy

wan-2-2/fun-camera

wan-2-2/fun-camera

Controls how strongly the camera motion is applied over time (higher = faster motion).
Number of denoising iterations; more steps refine detail and stability but take longer.
Controls how strongly the output adheres to the prompt versus allowing creative variation.
Offsets the diffusion sampling schedule, trading stability for stronger motion/style as the value increases.

Introduction of Wan 2.2 Fun Camera

Wan 2.2 Fun Camera transforms a single still image into a dynamic video with smooth pans, zooms, and rotations. Powered by the Wan2.2 family of models, it emphasizes cinematic camera motion without manual keyframes. You gain fast, repeatable results suited for social clips, hero shots, or product animations.

Wan 2.2 Fun Camera generates engaging video sequences from one source image, optimized for creators who need speed, consistency, or storytelling energy. It is ideal for artists, content makers, and product showcases, delivering MP4 outputs with clean, cinematic motion while preserving subject clarity.

Key Models for Wan 2.2 Fun Camera

Wan 2.2 Fun Camera 14B high-noise UNet fp8 scaled

This model establishes the early structure and motion dynamics from your still image. It drives the first diffusion steps that define global camera movement and motion richness. You can explore details of this model in the Hugging Face file.

Wan 2.2 Fun Camera 14B high/low-noise UNet fp8 scaled

This model refines and stabilizes frames after initial motion has been formed. It enhances details and ensures temporal consistency in the final animated output. More specifics can be found in the Hugging Face file.

Wan2.2 Image-to-Video LightX2V 4 Steps LoRA

The LightX2V LoRA variant accelerates the sampling process, offering faster iterations while slightly reducing motion complexity. Both high-noise and low-noise versions are available for flexibility. You can view the High-noise LoRA and Low-noise LoRA.

How to Use Wan 2.2 Fun Camera

Inputs Required

You must provide an image through the input labeled Image. This serves as the starting point for generating video. Additionally, you need to enter descriptive text in the Prompt input, which guides the subject intent and motion flavor. These are essential for building meaningful animated results.

Optional Inputs and Controls

You can adjust Width (px), Height (px), and Number of Frames by configuring these values, which control resolution and length of the output clip. There is also a Shift input available, allowing fine movement adjustment if needed. These settings let you balance output quality, duration, and creative motion.

Outputs

The result will be assembled as an MP4 video file. According to the documentation, the export defaults to a widely compatible H.264 format at a modest frame rate, making it suitable for fast previews and iteration. Your output will preserve the subject while animating the camera path you defined.

Best Practices

Start with clear, well-composed images when using the Image input to ensure best animation results. Keep prompts concise and action-oriented when filling in the Prompt input. For longer sequences or different aspect ratios, modify Width (px), Height (px), and Number of Frames carefully to prevent cropping and preserve motion balance.

Related Playgrounds