Wan2.2 Text-to-Image ComfyUI Workflow: Generate High-Quality Images Locally

About This Workflow

Generate high-quality images from detailed text prompts, including realistic people, cinematic scenes, abstract art, and stylized concepts.

Produce sharp textures and fine visual details such as skin, fabric, lighting, reflections, and environmental depth.

Handle both artistic and photorealistic styles, making it useful for creative art, concept design, and professional visuals.

Allow full local control through ComfyUI, letting you adjust steps, resolution, guidance, and LoRA strength for different results.

Run entirely on your own GPU, giving you privacy, unlimited generations, and faster experimentation without cloud limits.

Required Models

Diffusion Model

• wan2.2_t2v_high_noise_14B_fp8_scaled.safetensors
• wan2.2_t2v_low_noise_14B_fp8_scaled.safetensors

VAE

• wan_2.1_vae.safetensors

Text Encoder

• umt5_xxl_fp8_e4m3fn_scaled.safetensors

File save location

ComfyUI/
├───📂 models/
│   ├───📂 diffusion_models/
│   │   ├─── wan2.2_t2v_low_noise_14B_fp8_scaled.safetensors
│   │   └─── wan2.2_t2v_high_noise_14B_fp8_scaled.safetensors
│   ├───📂 text_encoders/
│   │   └─── umt5_xxl_fp8_e4m3fn_scaled.safetensors 
│   └───📂 vae/
│       └── wan_2.1_vae.safetensors

Workflow Graph