modelle/Jib Mix Wan - v2 WAN 2.2 Merge

Jib Mix Wan - v2 WAN 2.2 Merge

J1B

8/29/2025

9:47:06 AM

| Discussion|

Empfohlene Prompts

highly detailed

Empfohlene Negative Prompts

LightX lora with a negative weight

(bright:1.2), (overexposed:1.3), (high key:1.2), (blurry:1.2), (out of focus:1.2), (deformed:1.2), (ugly:1.2), (disfigured:1.2), (mutated:1.2), (extra limbs:1.3), (missing limbs:1.3), (bad anatomy:1.3), (malformed hands:1.3), (fewer digits:1.3), (extra digits:1.3), (low quality:1.2), (bad quality:1.2), (poorly drawn:1.2), (render:1.2), (cgi:1.2), (illustration:1.2), (drawing:1.2), (painting:1.2), (sketch:1.2), (cartoon:1.2), (anime:1.2), (watermark:1.2), (text:1.2), (signature:1.2)

Empfohlene Parameter

samplers

Res_3s, res_3s

steps

2 - 10

cfg

Tipps

Using LightX lora with negative weight increases realism but may require more steps.

Best sampler is Res_3s with Bong_Tangent scheduler from RES4LYF custom nodes.

Model can generate decent images/videos at as low as 6 steps due to speed loras mixed in.

Versions-Highlights

A merge of Wan 2.2 and My V1 mix

Ersteller-Sponsoren

I am using this great free txt2img workflow: https://pastebin.com/GPYQjUrx from AItrepreneur. Check out his Patreon: https://www.patreon.com/c/aitrepreneur/posts

RES4LYF custom nodes pack: https://github.com/ClownsharkBatwing/RES4LYF

Merge of my favourite Wan image enhancing loras, makes it much faster to use than when adding the lora separately.

Samplers

In my testing/research, I have found the best Sampler is Res_3s with Bong_Tangent Scheduler both from the RES4LYF custom nodes pack: https://github.com/ClownsharkBatwing/RES4LYF

I am using this great free txt2img workflow: https://pastebin.com/GPYQjUrx from AItrepreneur check out his Patreon https://www.patreon.com/c/aitrepreneur/posts

V2 is WAN 2.2. Merge that can do incredibly low steps down to 2 steps.
4 steps look better and 8 steps is great quality.

If you want a bit more realism, you can use the LightX lora with a negative weight, but you might have to then increase steps.

Can make decent images/videos in as low as 6 steps as it has speed loras mixed in.