Wan 2.2 Image-to-Video 14B GGUF | Beginner ComfyUI Workflow

Use this beginner ComfyUI workflow to turn an input image and text prompt into a video with Wan 2.2 Image-to-Video 14B GGUF.

The workflow loads and resizes an image, encodes text and vision inputs, runs Wan image-to-video processing, decodes the result, and combines it into a video.

LoadImage supplies the image, while CLIPTextEncode and CLIPVisionEncode handle the text and vision inputs.

WanImageToVideo performs the image-to-video step, and VAEDecode plus VHS_VideoCombine produce the video result.

Before configuring the workflow, review the usage video on its Civitai page.

Suggestion · not verified

The listed base model is Wan Video 2.2 I2V-A14B.

The workflow lists these model files: clip_vision_h.safetensors, lightx2v_I2V_14B_480p_cfg_step_distill_rank64_bf16.safetensors, umt5_xxl_fp16.safetensors, wan_2.1_vae.safetensors, wan2.2_i2v_high_noise_14B_Q5_K_S.gguf, and wan2.2_i2v_low_noise_14B_Q5_K_S.gguf.

It also lists these node packs: comfyui-gguf, comfyui-kjnodes, and comfyui-videohelpersuite.

The listed pin for comfyui-kjnodes is e2ce0843d1183aea86ce6a1617426f492dcdc802, and the listed pin for comfyui-videohelpersuite is 08e8df15db24da292d4b7f943c460dc2ab442b24.

The listed pin for comfyui-gguf is 1.1.4.

Before setup, verify which package provides comfyui-gguf and confirm the exact lightx2v_I2V_14B_480p_cfg_step_distill_rank64_bf16.safetensors file; its listed SHA-256 is 85c4a61c30e0497aa44b91d93a893b624708461a56fe5485183b28fa07e2dfb3.

Suggestion · not verified

No license is listed, so check the Civitai page before reusing the workflow.

Suggestion · not verified