If you like my work, please watch me on DeviantArt: https://www.deviantart.com/cloud-artisan, I post more there.
I highly suggest using a new Wan Fusion model. It gives much better results in just 10 steps. Here are the workflows: https://civarchive.com/models/1663553
Example prompt
A beautiful caucasian woman is bound and gagged on a chair (chairtied) in a bedroom with her legs spread extremely wide. Her hands are bound behind her back. Her athletic physique is restrained by ropes binding her upper arms and breasts. She wears tight white sports shorts and nothing else. Her long hair falls down her back like a waterfall, contrasting with massive cloth gag compressing her lips and pinching her cheeks. She intensely twists and turns to get free, but fails and remains tied up.Description
FAQ
Comments (12)
Wan is definitely better in terms of understanding what is expected of him. Definitely less deformation or some other things. I look forward to missionary sex😆.
Ok, ok. How we doing this.. How is the training done and how much vram do i need. I need info man.
How's WAN doing when it comes to following directions? As in something like, woman sits in a chair then picks up glass of water and takes a sip?
Hunyuan usually gets some of the steps right, but out of order or doesn't end the scene were it's told.
could u share the WF,thx
how do you find WAN compared to Hunyuan Video for this concept?
There are no workflows in the vids, and the workflow that you posted is a zipfile that does not contain a workflow but a link to some stupid utube video
Any chance for a 1.3B model version for us poor plebs ?
i also cannot getg a workflow from this vids... i download them from site by (right click save as ect)
put them in comfy and get >
Unable to find workflow in 2702272d-976c-4c93-8883-fc9a33c87ef2.mp4
can this lora handling i2v ?
Is this LORA just impossible to find by search anymore?
Has anyone tried this with image-to-video? I think it might be good for images with the right starter pic, since the characters stay well-tied vs normal image-to-video where the prompts of 'tied, bound' etc don't always hold.
hi you can update for wan 2.2?
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.