Unlock xxx images using flux, Qwen or Wan 2.2.
ATTENTION
Second version for WAN 2.2 T2V works with the usual models BUT for T2V I recommend some different models and LoRas:
https://huggingface.co/lightx2v/Wan2.2-Lightning/blob/main/Wan2.2-T2V-A14B-4steps-250928-dyno/Wan2.2-T2V-A14B-4steps-250928-dyno-high-lightx2v.safetensors instead of wan 2.2 high
https://huggingface.co/eddy1111111/WAN22.XX_Palingenesis/blob/main/WAN22.XX_Palingenesis_low_t2v.safetensorsinstead of wan 2.2 lowRemoved this recommendationNO speed/low step loras on high noise phase
https://huggingface.co/alibaba-pai/Wan2.2-Fun-Reward-LoRAs/blob/main/Wan2.2-Fun-A14B-InP-low-noise-HPS2.1.safetensors with STR:1.0 on low noise phase
https://huggingface.co/lightx2v/Wan2.2-Lightning/blob/main/Wan2.2-T2V-A14B-4steps-lora-250928/low_noise_model.safetensors with STR:1.0 on low noise phase
???
Profit!
All my videos have workflow embedded, the workflow is just a suggestion. But it works, and works very well.
First version for KREA.
Start using: 20 steps, res_2s, bong_tangent, STR 0.8
After that you can experiment with other arguments. It is not very good and deform hands and feet very badly.
ALL PICTURES ARE PNG WITH WORKFLOW INCLUDED
It is not perfect, sometimes it creates monsters, detached members and strange things, just try another seed or adjust your prompt. Eventually it get it right.
Examples:
Recommended Usage:
Checkpoint: flux1-dev
Strength: 1 (adjust depending on the image)
Sampling Method: DPMPP_2M
Scheduler Type: BETA
CFG Scale: 4
Steps: 35
Feel free to share your creations and feedback so I can improve this model and continue contributing to the creative community!
Description
This version shows a significant improvement in movement — it’s much more fluid and capable, able to simulate a wide range of motions that were previously impossible.
FAQ
Comments (37)
homie just comes in and drops this for WAN? GOAT
💦
Does this lora work for only T2V or can be used in IV2?
I don't know. It was trained for T2V, if you test please let me know your results.
@alcaitiff can confirm WAN V2 Lora works as well with I2V
@tremolo28 Nice! Thank you.
Published I2V version
You connect a positive prompt to WanVideoNAG while it is supposed to be a negative one. Uh?
I did?
Ops.
And it worked? WHY?
My entire world is a lie?
I've heard it works even better that way. Testing now
Maybe dumb. But didnt see workflow after downloading video and dragging to comfyui.
I’m sorry to hear that. I’m not sure what went wrong in your case, but I just downloaded the first video and tested it — it worked fine. The workflow is definitely embedded.
Have you tried experimenting with any of Eddy's other loras such as the Kinesis lora (which I believe is intended to replace the Lightx2v low lora), the FullDynamic_Ultimate_Fusion lora (no idea...) and the new Lightx2v WAIT lora... I'm not sure exactly what these do as there is no model page or descriptive readme...
No, you did?
I'm currently trying some combinations out... If I find one that seems to work really well I will let you know!
@Seeker360 thank you very much
I did try several of his LoRAs and couldn't produce anything I liked. I've since removed all of his stuff as there seems to be a lot of shadiness around them, not least of all his Sage Attention 3 wheels which overwrite a bunch of critical Python and PyTorch dependencies... From your changed recommendation, I assume you've come to a similar conclusion?
@Seeker360 yeah, I want to be out of the drama around him. And there are strange things but I'll not investigate. It's better to stay with things I know and trust.
does your suggested models works with i2v ? or if not do you have a suggestion for best config ? thanks a lot :) great work ! (tested with my workflows/models for now, it's really cool !)
@alcaitiff i did not found the dyno version for i2v, currently using the seko one, will stay with this then
Published the I2V version
steps, sampler, cfg etc for wan 2.2?
Drop any video on comfyui, they all have workflow with full config inside.
i cannot get this to work for qwen , im using qwen edit.
Same here, a strength 1 this gives only noise. At lower strength nothing is changed.
@derispan6661071 I’m sorry, but I haven’t provided any Qwen Edit or Qwen Edit Plus versions. The only Qwen version available is Qwen Image.
@alcaitiff No, no, dont be sorry. You're awesome and you make awesome job with your loras.
Nobody said it was a Qwen Edit model.
@javbukkake a lot of models for Qwen work on Qwen Edit, this one in particular I got the same result pure noise! but yeah it doesn't work! I was excited for Qwen Edit too! but thanks to the OP for the model it's fucking amzing for Qwen Image!
Same here, complete noise.
@henk717g qwen edit 2509 is cooking 160k steps in and still not good
@alcaitiff you're hero, you know that?
qwen_edit2509 cannot be used
I’m sorry, but I haven’t provided any Qwen Edit or Qwen Edit Plus versions. The only Qwen version available is Qwen Image.
@alcaitiff are you planning on it?
Yeah, it would be awesome. Qwen edit really needs a good NSFW lora
Details
Files
Wan2.2-High-MysticXXX-v2.safetensors
Mirrors
w22_t2v_MysticXXX-v2_H.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-Mystic-v2.safetensors
T2V_MysticXXX_HIGH.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-t2v-High-MysticXXX-v2.safetensors
General-NSFW-MysticXXX-Wan Video 2.2 T2V-A14B_HIGH.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors
Wan2.2-High-MysticXXX-v2.safetensors