You can click on the link below and try it out directly. If the effect is good, you can deploy it locally.
Online Workflow: Wan2.2 Remix i2v 3.0 Image-to-Video Automatic Reverse
Experience link: https://www.runninghub.ai/post/2033536710499373058/?inviteCode=rh-v1325
Online Workflow: Wan2.2 Remix I2V v2.1 SVI2Pro Fully Automatic Reverse Long Video
Experience link: https://www.runninghub.ai/post/2008833586203463682/?inviteCode=rh-v1325
Online Workflow: Wan2.2-Remix-T2V-v2.0 text-to-video workflow
Experience here: https://www.runninghub.ai/post/1991533843182264322/?inviteCode=rh-v1325
Online Workflow: Wan2.2-Remix-I2V-Comfy-Qwen3
Experience here: https://www.runninghub.ai/post/1986632318448267265/?inviteCode=rh-v1325
Now! Register to get 1000 points, daily login 100 points, play 4090! Experience the super power of 48G!
[B-site video]: https://www.bilibili.com/video/BV1hF2cBNECZ
[Hugging Face: Wan2.2-Remix]:https://huggingface.co/FX-FeiHou/wan2.2-remix
[Telegram Group]: https://t.me/wan22remix
The workflow is very simple to use, and I have uploaded a tutorial on how to use it on YouTube. You can try it out and follow the tutorial and Civitai's workflow to easily reproduce the results. Thank you. If you have any other questions, please leave a comment in the comment section
Model Overview — Wan2.2-Remix
Wan2.2-Remix is a research and creative model built to generate short, cinematic video clips directly from text prompts. It focuses on human dynamics, realistic motion, and scene consistency, producing lifelike movement and smooth transitions without requiring LoRA setup.
This model already includes NSFW capabilities, enabling the generation of high-quality NSFW-themed videos without the need for additional LoRA.
High-noise model T2V: Lightx2v Wan2.2 Lightning dyno version (quantized to fp8)
High-noise model I2V: Comfy-Org Wan 2.2 ComfyUI Repackaged (fp16, quantized to fp8)
Low-noise model: Comfy-Org Wan 2.2 ComfyUI Repackaged (quantized to fp8)
Enhancements: Blended with open-source motion LoRA data and refined pose training for enhanced realism, anatomical precision, and gesture diversity.
Integration: Fully compatible with
lightx2v, no extra setup required.Recommended CLIP: NSFW-Wan-UMT5-XXL
Recommended Settings:
Steps: 4–8
CFG: 1
Shift: 5–10
Sampler: Euler
Scheduler: Simple
This release is currently beta. While functional, it is not yet perfect—feedback and testing are highly appreciated.
The GGUF version of the Remix model is now available.
Thanks to @BigDannyPt for creating it!
Download link:Wan2.2-Remix I2V GGUF v2.0
Changelog
I2V v3.0
Added VBVR LoRA to enhance spatial understanding and improve scene consistency.
Introduced UnifiedReward optimization to improve overall video quality and generation stability.
Adjusted several NSFW LoRAs to improve quality and visual consistency in NSFW scenes.
This version is considered the final release of the Remix I2V series.Unless there are major technological breakthroughs in the future, no further updates are planned.
I2V 2.1
Enhance facial consistency in long videos
Adjust LoRA ratio to make video dynamics more realistic
The v2.1 version has reduced the weight of the NSFW in the low-noise model to enhance character consistency. If you want to get a better NSFW, you can mix and match the low noise model of version 2.0.
T2V 2.0
Optimized character stability in generated videos, reducing jitter and unwanted motion artifacts
Enhanced video detail for more visually rich scenes
Added dynamic motion effects to improve realism and cinematic impact
Updated base model to original Wan2.2 version to minimize differences across random seeds
I2V 2.0
Significantly enhanced character motion and poses for more dynamic animations
Greatly improved camera movement and cinematic framing
Optimized NSFW content weighting to reduce jitter in generated videos when producing SFW scenes
Enhanced explosions and particle effects for more impactful visuals
SFW version will no longer receive updates
I2V 0.8
Updated lighting acceleration module to Wan2.1 version, fixing slow-motion character issues
Integrated over 20 open-source LoRAs and 10 custom-trained LoRAs to enhance motion dynamics
Removed several low-quality LoRAs to improve overall visual fidelity
T2V 1.0
Reduced dyno weighting to improve stability across different random seeds
Added more Asian facial datasets for greater diversity
Enhanced character motion dynamics for fluid animation
Improved motion amplitude, gesture diversity, and overall visual quality
T2V 0.9
High-noise T2V model: Lightx2v Wan2.2 Lightning dyno version (fp8)
High-noise I2V model: Comfy-Org Wan 2.2 ComfyUI Repackaged (fp16, fp8 quantized)
Low-noise model: Comfy-Org Wan 2.2 ComfyUI Repackaged
Refined character motion blending for smoother transitions
Enhanced anatomical articulation and natural body flow
This model incorporates the following Loras. Thank you to the Lora authors for their work. In order to facilitate management, I will change the file name uniformly after downloading Loras, so I cannot correspond to their real names one by one. Please understand. If there is any infringement, please contact me and the next version will remove it.
Wan2.1 - lightx2v_I2V_14B_480p_cfg_step_distill_rank256_bf16
Wan2.2 - I2V_A14b_high_noise_lora_rank64_lightx2v_4step_1022
Wan2.2 - I2V_A14b_low_noise_lora_rank64_lightx2v_4step_1022
Wan2.2 - Fun-A14B-InP-HIGH-MPS_resized_dynamic_avg_rank_21_bf16
Wan2.2 - Fun-A14B-nP-LOW-HPS2.1 resized dynamic avg rank 15 bf16
Wan2.1 - Skyreels-v2-T2V-720P LoRA rank64 fp16
Wan2.2 - PusaV1_lora_HIGH_resized_dynamic_avg_rank_98_bf16
Wan2.2 - PusaV1_lora_LOW_resized_dynamic_avg_rank_98_bf16
Wan2.1-T2V-BBFW - Better Buttholes For WANv1
Wan2.1-12V-General NSFW-Low
Wan2.2-12V-Hip Slammin Assertive Cowgirl
Wan2.2-12V-PENIS LORA e496
Wan2.2-12V-DR34ML4Y -All-In-One NSFW
Wan2.2-T2V-Body Shots v2 14B
Wan2.2-12V-Deepthroat, Blowjob
Wan2.2-T2V-Masturbating Rubbing 14B
Wan2.2-T2V-Doggy Style-14B -v3
Wan2.2-12V-COWGIRL+REVERSE COWGIRL-14b
Wan2.2-12V-Combo Handjob Blowiobsafetensors
Wan2.2-Bounce-Test
Wan2.2-T2V-GodPussy
Wan2.2-T2V-MysticXXX
My Train:
P001-Side-Lying-Sex-Wan2.2-i2v v10
P003-Sexy-Enhancer-Wan2.2-i2v
P004-Lying-Bouncing-Wan2.2-i2v v20
P005-Chinese-Dance-i2v-v20
P006-Squirting-Wan2.2-i2v
P008-Running-Wan2.2-i2v v10
P009-The-Walk-Wan2.2-i2v-Light
P010-whip-t2v-v10
F006-AsianGirls-Wan2.2-t2v v10Disclaimer
This model is released for research and creative purposes only. It must not be used for illegal, unethical, or harmful activities. The author assumes no responsibility for any misuse or consequences arising from its use.
Description
FAQ
Comments (62)
what settings can i change to remove the moles/black spots that shows on skin in the output video?
Some prompt like fair skin or smooth skin could help. Let me know if it worked! Cheers 🙌
Amazing checkpoint! This is pretty much the only checkpoint that I have had good results with apart from the base model!
Do you plan on releasing fp16 versions of the I2V 3.0 model?
Hello, when will the GGUF version of remix V3.0 be released?
@jimzlf Thank you very much for the reminder.
Any hope to get BF16 or FP16 so we can make mxfp8 and INT8 quants properly? 😁
I think the most noticeable difference between Version 3.0 and Version 2.1 is that the pace of the action has slowed down. For example, the intensity of actions such as foot play, oral sex, and penetration isn’t as intense as it was in the previous version. In other words, the action in the videos is more steady, and the attention to detail may be more refined. This is just my personal opinion, for your reference only.
but muh intense penetration!!!!11
That one, and the fact that 3.0 might've been trained on BigBreasts, all of my outputs have come out with large breasts
Do you perhap know how to use the reverse prompt node in his workflow local? need to get runninghub membership or something?
In general, in my opinion this model has gone backwards since 2.0 if you use it for NSFW. If people are considering this model for NSFW generations, use 2.0 with genital helper loras. It creates much more believable scenes with good motion, better prompt adherence too.
@yannickm wich genital helper lora to use?
probably because of VBVR LoRAs. it slows down the motion, albeit with more details refined. if its not baked, you can generate more motion with negative strength on the VBVR LoRAs.
bad results. i swapped the most base/standard wan2.2 models (high and low) with the remix v2 models, and my output looks horrible
Haven't used this model, but I'm going to guess it has lightning loras baked in. If your old workflow used lightx2v or similar, try disabling that. If it did not use those, make sure to check that cfg is 1 and steps in the 4-8 range.
use author recommended workflow. ngl, but its probably the best basic wan2.2 i ever had
This is the best nsfw wan ever made... Check your WF.
@raveGoingDark that workflow has a runninghub API node. Could you tell me how to use it on local comfyui?
@nikijiki0402 check description, the author attached huggingface link. thats where to find some local wf along with their repo model
lightx2v
yeah, don't make my mistake of deleting huge amounts of GBs because you thought a model was bad.. normally it's all a matter of seeing it it has lighting loras baked in, or using a different CFG and steps, always make sure to read liner notes on these things.
@raveGoingDark ty. I've checked those but doesnt look like it has the same prompt reverse like online wworkflow
Will pay yellow buzz for t2v v.3 😄. Just saying 😎
Dunno how much you paid but thanks! 🤣
Best model ever. You can toss out 90% of your lora mess and everything works perfect. I highly advise using Painter node for dynamic motion.
I personally like that these models don't have identical faces, unlike Smooth Mix, where the same face is constantly generated.
Как понимаю lightx2v встроенный?
или его необходимо подключать?
Finally a model and workflow that actually work. Thank you :)
Anyone with a working workflow? The one I downloaded from the Youtube video has missing nodes which cannot be installed.
download it from comfyUI manager, 1- open manager. 2-update comfyui. 3- install missing nodes. These are buttons inside manager. Why aren't you using chatgpt, google to solve your problems . Everyone has same problem after copypasting .jason files
@oulu1619 protip: comfyui manager doesn't actually install every missing node everytime
@DirtyRowdyJackerBoy my desktop version of Comfyui couldn't find the missing node. The Swarmui version of Comfyui was working fine until I updated it then Wan 2.2 I2V stopped working. Too many clashes after updating Comfyui
Thanks for your awesome work.
For Wan2.2-Remix-I2V-Comfy-Qwen3 workflow, that QWEN 3 VQA node refuses to see my Qwen Abilerated models. My other QWEN VL models see the. I know they modified the model location by providing a HF_models and GGUF_models json and some use the custom_model.json. No matter where I update it, it doesn't show the abilerated models in the QWEN 3 VQA node, any idea?
Otherwise the workflow isn't NSFW from what I am getting despite the excellent quality.
I'll need to try a different workflow. Results aren't terrible but getting smoothmix issues. I think mixing Wan causes it to lose a few concepts. Can't do anything too much out of the ordinary, then seems to mutate the characters. They also seem to turn asian every so generation.
Maybe a different workflow is required from my standard simple light weight Wan2.2 I2V workflow. But still thanx for the model! Appreciate anyone doing the task of trying to evolve the models.
3.0 workflower error:
"openai.APIConnectionError: Connection error."
But my internet connection is working fine.
3.0 workflow differs from the workflows on YouTube and RunningHub; RunningHub uses an API, and an API connection error occurred during execution.
best mod i have used,
Great model. I have two issues tho:
1) The process will sometimes stop at first KSample node around 67% for some reason and then it stopped using my GPU (5070)
2) Sometimes it will say API connection error
Great workflow. Clean and easy to use, no bloat. Highly recommend combining it with LoRAs for the best results
the result is really better than the previous version
hello is it possible for you to uplaod the t2v 2.0 in bf16 ?
I'll give you my father and mother if you want them xD
with the latest v3.0 using the same prompt, i cant seem to generate small breasts and keep getting large breasts
Quality is great. Several others have some quality issues due to the loras that are baked into them, making things blurry. I don't get that here. Anyway to control the humping?
The only thing I haven't figured out is how to make someone not actively hump in a sexual activity. Everyone seems to want to hump. Cowgirl on motionless dude, bros thrusting his bits. Woman is still while..., she's totally not still. Lol. Any tips?
No mater how much i prompt it the girl breasts size always big, even when the starting image already shown she has small size breasts.
Try breast size LoRAs, one example is Breast Slider from ZIT.
I dont know why people praising so much this model, like the breasts always change to big size breasts and everyone in the frame humping air for no reason.
Works well for me. Good prompt control without lora's.
Is this model better than wan22EnhancedNSFWSVICamera_nolightningSVICfQ8 ??
I just switched from that SVI model, to this one, and I'm getting a lot less flicker with this model (I'm using the Q5 version).
is there a gore version?.
why still v3 is not good for nsfw? v2.0 is good for it. I use v3 high and v2.0 low for nsfw. so v3 features is in high or am I wasting using the v3 high?
Does anyone have a flf2v workflow for this version? I'm having trouble applying the existing process to this version.
Do you need to trigger the incorporated LORA's ?
3.0 feminizes all characters that appear. Is anyone else experiencing the same issue as me?
Yeah and kinda ruins likeness which is a shame because the movement is quite a bit better than most. In 2.0 anyway.
Why do all videos with this model in gallery look like slowmo? I thought it's just me but I didn't have this problem with other models.
can anyone share a workflow with continuous generation and not just 5 second clips? would really appreciate it
https://civitai.red/models/1823416/wan-22-image-to-video-with-caption-and-postprocessing?modelVersionId=2558117&sync-account=green
I am having good results with the multi clip long look one. I do not use the one with ollama integrated as it's been taking way too much time and can often return errors. I use ollama on its own to generate the 4 prompts then copy paste them in the workflow. Just make sure to toggle the red nodes below the individual clip nodes to FALSE
@9aouad doesnt that workflow take unet GGUF? or do you manuallyswpa them out for regular model loaders?
@xx_debby349 I use it with the GGUFs yep but I don't see why it wouldnt work with the full models.
Edit : The author said it works in one of the comments so you're good
v3 can certainly do things that other models can't. however it has biases towards poses: it heavily forces rear view, even if that means bend and rotate characters unrealistically. I tried to prompt front view in different ways, yet I had that issue many times with different generations. check your captions
Details
Files
wan22RemixT2VI2V_i2vHighV30.safetensors
Mirrors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
wan22RemixT2VI2V_i2vHighV30.safetensors
GonzalezHigh.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_v3.0.safetensors
Wan_Base_HI.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
wan22RemixT2VI2V_i2vHighV30.safetensors
wan22RemixT2VI2V_i2vHighV30.safetensors
Wan22_Remix_NSFW_i2v_14b_high_lighting_v30.safetensors
wan22RemixT2VI2V_i2vHighV30.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
wan22RemixT2VI2V_i2vHighV30.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
Wan22_Remix_NSFW_i2v_14b_high_lighting_v30.safetensors
Remix NSFW I2V Lighting E4M3FN v3.0 FP8 HIGH - Wan2.2.safetensors
wan22RemixT2VI2V_i2vHighV30.safetensors
wan22RemixT2VI2V_i2vHighV30.safetensors
Wan2.2_Remix_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors
wan22RemixT2VI2V_i2vHighV30.safetensors
Wan2.2_NSFW_i2v_14b_high_lighting_fp8_e4m3fn_v3.0.safetensors