Wan 2.2 Text To Video (T2V)
Video Lecture
| Section |
Video Links |
| Wan 2.2 Text To Video |
  |
Description
We will use the GGUF quantised Wan2.2 models.
📂 ComfyUI/
├── 📂 models/
│ ├── 📂 clip/
│ │ └── umt5-xxl-encoder-Q8_0.gguf
│ ├── 📂 loras/
│ │ ├── Wan2.2-T2V-A14B-lora-high_noise.safetensors
│ │ └── Wan2.2-T2V-A14B-lora-low_noise.safetensors
│ ├── 📂 unet/
│ │ ├── Wan2.2-T2V-A14B-HighNoise-Q8_0.gguf
│ │ └── Wan2.2-T2V-A14B-LowNoise-Q8_0.gguf
│ └── 📂 vae/
│ └── wan2.1_vae.safetensors
Sample Prompts
| Prompt |
Workflow |
| A lone fisherman rows across a foggy bay at dawn. The fisherman wears a weathered coat, his face lined with years of sea life, and a wooden oar creaks against the boat. The camera begins with a wide aerial shot, then slowly cranes down to eye level as ripples widen across the water. Lighting is soft, bluish tones of early morning, with golden sunlight breaking faintly through the mist. |
 |
| A man walks across a desert road with a suitcase as heat waves shimmer in the distance. He wears a dusty suit, sweat-stained collar, each step heavy against the endless expanse of cracked asphalt and sand. The camera follows in a long tracking shot from behind, lens slightly warped by rising heat mirages. Lighting is harsh midday sun, stark shadows, with muted desert browns and pale skies. |
 |
| A child releases a paper lantern into the night sky during a village festival. The lantern drifts upward, joining hundreds of glowing lights rising above rooftops and into the dark expanse. The camera tilts up with the lantern, then cranes higher to reveal the night sky filled with floating orbs. Lighting is warm golden tones from lanterns, balanced against cool starlight above. |
 |
| A surfer waits on their board, floating in still waters just before dawn. The ocean is glassy and calm, horizon glowing faint orange as seabirds pass overhead, the surfer poised in anticipation. The camera starts beneath the waterline, half submerged, then tilts up to reveal the silhouette against the sunrise. Lighting is soft pre-dawn hues, with gradients of lavender and pale pink. |
 |
With Scene Shifts
| Prompt |
Workflow |
| A woman sips coffee at a small cafe table on a quiet street corner in autumn. She wears a wool scarf, notebook open beside her cup, leaves drifting down from nearby trees. One second in, a gust of wind flips the notebook pages and sends a loose leaf skittering across the cobblestones. The camera begins on a medium shot of her sipping, then pans quickly to track the runaway paper. Lighting is soft golden morning light with warm amber and russet tones from fallen leaves. |
 |
| A young man plays guitar on a rooftop overlooking a city skyline at dusk. His jacket is draped beside him, neon lights beginning to glow far below as he strums gently. Two seconds in, a flock of pigeons suddenly bursts upward from the roof edge, scattering into the fading sky. The camera starts with a wide shot of the rooftop, then tilts up to follow the rising birds. Lighting is deep violet and orange twilight, with faint city neon flickering below. |
 |
| A woman practices yoga in a quiet living room, calmly holding a balance pose; two seconds in, her foot slips slightly on the mat and the camera shifts to her hands bracing on the floor. The camera starts wide from the front, then pushes into a low angle on her hands/feet. Lighting is soft window light, morning brightness with gentle shadows. |
 |
| Two friends sit on a train, one gazing out the window while the other sketches in a notebook. The window shows fields rushing by, reflections flickering across their faces. One second in, the sketcher's pencil snaps, startling both of them. The camera begins focused on the friend staring out the window, then quickly pans to the sketcher fumbling with the broken pencil. Lighting is soft morning light filtering through the moving train windows, casting shifting shadows. |
 |
Wan2.2 Video Dimensions
landscape
768x432 (16/9)
832x480 ~(16/9)
848x480 ~(16/9)
768x512 (4/3)
910x512 (16/9)
1024x576 (16/9)
1280x720 (16/9)
1424x800 ~(16/9)
1824x1024 ~(16/9)
1920x1080 (16/9)
portrait
272x480 ~(9/16)
288x512 (9/16)
480x848 ~(9/16)
480x644 ~(9/16)
512x768 (3/4)
576x1024 (9/16)
720x1280 (9/16)
800x1424 ~(9/16)
1024x1824 ~(9/16)
1080x1920 (9/16)
WGET Commands
If you are using Runpod, or a similar hosted GPU service, then you can access your running pod/instance using a terminal.
| # CD into ./ComfyUI/models/clip/ folder
wget https://huggingface.co/city96/umt5-xxl-encoder-gguf/resolve/main/umt5-xxl-encoder-Q8_0.gguf
|
| # CD into ./ComfyUI/models/loras/ folder
wget https://huggingface.co/lightx2v/Wan2.2-Lightning/resolve/main/Wan2.2-T2V-A14B-4steps-lora-rank64-Seko-V2.0/high_noise_model.safetensors -O Wan2.2-T2V-A14B-lora-high_noise.safetensors
wget https://huggingface.co/lightx2v/Wan2.2-Lightning/resolve/main/Wan2.2-T2V-A14B-4steps-lora-rank64-Seko-V2.0/low_noise_model.safetensors -O Wan2.2-T2V-A14B-lora-low_noise.safetensors
|
| # CD into ./ComfyUI/models/unet/ folder
wget https://huggingface.co/QuantStack/Wan2.2-T2V-A14B-GGUF/resolve/main/HighNoise/Wan2.2-T2V-A14B-HighNoise-Q8_0.gguf
wget https://huggingface.co/QuantStack/Wan2.2-T2V-A14B-GGUF/resolve/main/LowNoise/Wan2.2-T2V-A14B-LowNoise-Q8_0.gguf
|
| # CD into ./ComfyUI/models/vae/ folder
wget https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/vae/wan_2.1_vae.safetensors
|
Wait for files to download fully before running your workflows.