Wan2.2 Fun Camera turns a single still image into a lively video with smooth pans, zooms, and rotations. Built around the Wan 2.2 Fun family, this ComfyUI workflow focuses on camera movement that adds depth and story energy without manual keyframes or editing. If you need quick social clips, dynamic hero shots, or punchy motion for product and character stills, Wan2.2 Fun Camera delivers clean, repeatable results.
The graph ships with two variants so you can choose the right balance for your task. The native fp8 scaled path maximizes motion richness from the Wan2.2 Fun Camera models, while the LightX2V 4 Steps LoRA variant prioritizes speed with a small tradeoff in dynamics. Both variants share the same prompts, camera presets, and export stage, making Wan2.2 Fun Camera easy to learn and fast to iterate.
The workflow contains two parallel branches: “Wan2.2 Fun Camera fp8 scaled + LightX2V 4 Steps LoRA” and “Wan2.2 Fun Camera fp8 scaled.” Enable one branch and disable the other before you generate. Both branches follow the same four operational steps and a two-stage denoising schedule that starts with high noise and finishes with low noise for sharper frames.
This group prepares the Wan 2.2 Fun Camera high-noise and low-noise UNets, the optional LightX2V LoRA pair, the UMT5-XXL text encoder, and the Wan 2.1 VAE. Choose the LoRA branch when you need faster turnarounds and the native fp8 branch when motion richness is the priority. Once loaded, the models remain resident, so you can iterate on camera moves and prompts rapidly in Wan2.2 Fun Camera.
Add your source still in the “Upload start_image” group. The workflow expects a single image and will preserve core content while animating the viewpoint. For best results in Wan2.2 Fun Camera, start with a clean subject, decent contrast, and a composition that leaves room for movement at the edges.
Enter a concise positive prompt to describe subject intent and motion flavor, then fine-tune with a negative prompt to avoid artifacts. Keep prompts short and action-oriented, for example “cinematic product hero, gentle zoom out” or “portrait, subtle pan right.” Wan2.2 Fun Camera responds well to style and motion descriptors without over-constraining the scene.
Pick a camera preset and timeline in the WanCameraEmbedding group. This builds a camera path embedding plus duration and resolution settings that the core node will follow. Wan2.2 Fun Camera includes intuitive presets such as zoom out, pan, and rotate, so you can preview different motions quickly without hand-tuning curves.
The first sampler pass lays down coarse structure and initiates motion dynamics using the high-noise UNet. This is where the chosen camera path starts to manifest as global movement. In the LoRA branch, the LightX2V 4 Steps LoRA compresses this stage into fewer steps for speed. In the native fp8 branch, Wan2.2 Fun Camera prioritizes motion fidelity before refinement.
The second sampler pass focuses on detail recovery, texture, and temporal consistency using the low-noise UNet. It respects the established camera path while cleaning up edges and faces. With LoRA enabled, this stage also runs faster at a slight cost in micro-motion variation. Without LoRA, Wan2.2 Fun Camera squeezes extra crispness from the final frames.
The VAE decodes latents back to frames and the Video Helper node assembles them into an MP4. The default settings target a widely compatible H.264 file at a modest frame rate so previews render quickly. You can iterate in-place by adjusting prompts or camera presets and regenerating. Wan2.2 Fun Camera keeps your outputs consistent with predictable filenames for batch experiments.
WanCameraEmbedding (#87, #108)Defines the motion plan for Wan2.2 Fun Camera. Choose a camera preset and set resolution and clip length for the whole shot. If you see cropping or letterboxing, adjust width and height here. Longer clips require more compute and may benefit from the LoRA path for speed.
WanCameraImageToVideo (#80, #98)The core driver that fuses prompts, VAE, start image, and the camera embedding into motion-aware latents. Treat this as the bridge between your creative intent and the diffusion samplers. If motion feels too strong or too subtle, revisit the camera preset or prompt verbs rather than the sampler first.
KSamplerAdvanced (#71 and #78, #102 and #103)Runs a two-stage schedule: a high-noise sampler to establish motion, then a low-noise sampler to refine. For stronger structure use a few more steps in the first pass, for crisper details invest steps in the second pass. Keep guidance balanced so motion remains natural and does not overfit to the prompt.
VHS_VideoCombine (#116, #117)Packages decoded frames into a single video file for quick review. Adjust the frame rate to match the feel of your camera move and tune CRF if you need higher quality exports. Use a clean filename prefix to keep versions organized while exploring Wan2.2 Fun Camera variations.
LoraLoaderModelOnly (#88, #90)Enables the LightX2V 4 Steps LoRA pair. Turn these on for faster iterations or lower VRAM pressure. Disable them for maximum motion richness when finalizing a Wan2.2 Fun Camera shot.
This workflow implements and builds upon the following works and resources. We gratefully acknowledge Alibaba PAI for the Wan2.2-Fun A14B Camera Control model, Comfy Org for the ComfyUI Wan2.2 Fun Camera Control workflow and documentation, and aigc-apps for the VideoX-Fun framework for their contributions and maintenance. For authoritative details, please refer to the original documentation and repositories linked below.
Note: Use of the referenced models, datasets, and code is subject to the respective licenses and terms provided by their authors and maintainers.
RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Playground, enabling artists to harness the latest AI tools to create incredible art.