models/Lightx2v V2 Wan2.2 Vace - 3 steps - Wan2.2 Vace H/L Clean FP8

Lightx2v V2 Wan2.2 Vace - 3 steps - Wan2.2 Vace H/L Clean FP8

12/22/2025

5:06:35 PM

Related Keywords & Tags

base model,checkpoint,lightx2v v2 wan2.2 vace - 3 steps,omegawpn,stable diffusion,uni_pc,video lip sync model,wan video 14b t2v,wan2.2 vace h/l clean fp8,wanvacetovideo

Recommended Prompts

A woman is walking peacefully through a sunny park, holding a leash as her playful dog trots beside her.

Suddenly, the dog breaks free and dashes off— the woman gasps in surprise and reaches out, calling after it.

As she starts to chase the dog, she trips over a root and tumbles to the ground, startled and flustered.

Moments later, the dog turns around and runs back toward her, tail wagging, as if trying to help or comfort her.

Recommended Negative Prompts

色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走, 3d 渲染、视频游戏. unantural, 2d, anime, cartoon, unrealistic, fake, painting, blurry, low-resolution, low contrast, poorly rendered, illustration, low detail, low-res, deformed, washed-out, flat lighting, text artifacts.jerky movement, abrupt cuts.Visible seams, texture repetition, tiling artifacts.text, distortion, ugly, slow motion, slow movement, unnatural movement, stiff limbs, sliding feet, floaty animation, lack of weight, robotic animation, physics glitch

Recommended Parameters

samplers

Uni_pc, Uni_pc_bh2, UniPC

steps

3 - 7

cfg

resolution

416x720, 480x832, 576x1024

Tips

For vace lip sync, avoid using too many LoRA models as they can interfere with the preprocessor; it is preferred not to use any LoRA.

Shift parameter between 1 to 10 is great; this model was tested at shift 8.

Use Wanvacetovideo strength at 0.70 with a maximum of 81 frames when applying to video.

Version Highlights

This is a clean version of Wan2.2 High_noise, and Low_noise without any lora.
Don't forget to leave a like :)

Creator Sponsors

Omegawpn_vace_autoprompt_4-7 steps workflow - A recommended workflow model for enhanced lip sync results.

Only tested on Native wan workflow !!

[3/3 [01:03<00:00, 21.17s/it]

120 frames with 3 steps took me 60 sec using sageattn.

Steps: 3.

Cfg: 1.

Shift: 8 ( 1 to 10 is great, have not tested any higher)

Sampler_name: Uni_pc or Uni_pc_bh2. ( all of them work but Uni_pc is the best lol)

Scheduler: Normal, ( simple, beta or others work too)

For vace lip sync, do not use too many lora's, it will screw up the preprocessor. I prefer not to use any lora.

Wanvacetovideo Strength: 1.00.

Width: 416.

Height: 720.

Set wanvacetovideo strength to: 0.70 and use this prompt with max 81 frames. You can use any image of a woman, close-up or full body image, with or without a dog, it does not matter.

0, 24 "A woman is walking peacefully through a sunny park, holding a leash as her playful dog trots beside her."

24, 49 "Suddenly, the dog breaks free and dashes off— the woman gasps in surprise and reaches out, calling after it."

49, 61 "As she starts to chase the dog, she trips over a root and tumbles to the ground, startled and flustered."

61, 81 "Moments later, the dog turns around and runs back toward her, tail wagging, as if trying to help or comfort her."

You can use higher resolution such as 480x832 or 576x1024 but 416x720 has great quality too.

Omegawpn_vace_autoprompt_4-7 steps_workflow can be found here see link below.

https://civitai.com/models/1806036/omegawpnvaceautoprompt4-7-stepsworkflow

Credit goes to the original creators of the checkpoints. My only contribution was merging them.