Home Creators Posts Import Register Favorites Logout

Downloads

Content

🚨UPDATE 29 OCT 2025🚨 - The latest update of ComfyUI broke the cg-Everywhere nodes I am using in most of my newer workflows. I have got your back: https://www.patreon.com/posts/emergency-wan-2-142314304

Hey friends! 💛I’ve put together a dead-simple guide so you can get Wan 2.2 (A14B, GGUF) running without guesswork. You’ll need two models (HighNoise + LowNoise), a Text Interpreter (UMT5-XXL), and a VAE. Grab everything here:

💻Software:ComfyUI | Generate video, images, 3D, audio with AI🏎️ Sage Attention Install Guide:https://youtu.be/q4jhieMwNkUSage Attention One Click installer for my Patrons:🎁https://www.patreon.com/posts/edit-31-08-2025-135625892

👉 Repo:

Text 2 Videohttps://huggingface.co/QuantStack/Wan2.2-T2V-A14B-GGUF/tree/main

Image 2 Video

https://huggingface.co/QuantStack/Wan2.2-I2V-A14B-GGUF/tree/main

What to download (match to your GPU VRAM)

4 GB: 14B isn’t realistic. Use the 5B variant instead.

https://huggingface.co/QuantStack/Wan2.2-TI2V-5B-GGUF/tree/main

8 GB: Q3_K_S or Q3_K_M for both HighNoise + LowNoise. Keep the Text Interpreter on CPU.

12 GB: Q4_K_S (or Q4_1) for both. Text Interpreter Q5_K_M (CPU or GPU if you’ve got room). VAE can be the smaller one if you’re tight.

16 GB: Q5_K_S (or Q5_0 / Q5_1) for both. Text Interpreter Q5_K_M on GPU, Wan 2.2 VAE recommended.

24 GB: Q6_K for both. Interpreter on GPU, Wan 2.2 VAE.

32 GB: Q8_0 for both. Interpreter on GPU, Wan 2.2 VAE Or go for the Fp8 version from Kjai

Rule of thumb: pick the same quant for HighNoise + LowNoise. Start conservative (Q4/Q5) and step up if you’ve got headroom.

Download these add-ons

Text 2 Video Lroa (Lightning x2v, 4-step)

High-noise: wan2.2_t2v_lightx2v_4steps_lora_v1.1_high_noise.safetensors

Low-noise: wan2.2_t2v_lightx2v_4steps_lora_v1.1_low_noise.safetensors

Image 2 Video Lora (Lightning x2v, 4-step)

High-noise: wan2.2_I2v_lightx2v_4steps_lora_v1.1_high_noise.safetensors

Low-noise: wan2.2_I2v_lightx2v_4steps_lora_v1.1_low_noise.safetensors

VAE

wan_2.1_vae.safetensors

Text Encoder (UMT5-XXL)

umt5_xxl_fp8_e4m3fn_scaled.safetensors

LoRa's used:

Golden by T2Video:

https://civitai.com/models/1671285/retro-90s-anime-golden-boy-style-lora-wan-22-14b

WAN2.2_Spatial Magic—空间魔法:

https://civitai.com/models/1867349/wan22spatial-magic

Files

WAN Video 2.2 is here—This Uncensored Model is a game changer. 🚀

In this tutorial, I’ll show you how to unlock the full power of WAN’s latest upgrade: stunning higher-quality videos, smarter lighting and camera control, full LoRA support, and yes—it’s still 100% uncensored. You’ll learn how to: ✅ Generate videos from text prompts and images ✅ Supercharge your results with LoRAs ✅ Extend video length without breaking consistency ✅ Optimize performance for your GPU By the end, you’ll have everything set up in ComfyUI, ready to create cinematic, mind-blowing AI videos right from your local machine. Oh, and I’ve got bonus tips on frame interpolation and extending videos even further. 👉 Don’t forget to grab my ready-to-use workflows and VRAM guide on Patreon—everything neatly mapped out so you can focus on creating, not troubleshooting. This is the future of local AI video generation. Let’s dive in. 🎥✨ 📅 Book Your Personal 1-on-1 Session: 👉 Schedule Here: https://koalendar.com/e/1hr-1-on-1-stable-diffusion 🎁 Free Goodies & Essential Resources: • 🎯 Ultimate Beginner's Guide: https://www.patreon.com/posts/sneak-peek-alert-90799508 • 🛠️ Free Workflow Guide: https://www.patreon.com/posts/get-your-free-99183367 • 🎨 My Exclusive Style Collection: https://www.patreon.com/posts/my-collection-of-87325880 • 🚀 Support & Get Exclusive Content: 👉 Join Patreon Community: https://patreon.com/AIKnowledgeCentral 🎬 Popular Playlists: • 🤖 Invoke AI: Watch Here: https://www.youtube.com/watch?v=70HTvc-4gek&list=PL8HbYkSF5f9Ggts03f7ttPzFJyFiHgqoJ • ⚡ Flux: Watch Here: https://www.youtube.com/watch?v=70HTvc-4gek&list=PL8HbYkSF5f9Ggts03f7ttPzFJyFiHgqoJ • 🛠️ Forge: Watch Here: https://www.youtube.com/watch?v=70HTvc-4gek&list=PL8HbYkSF5f9Ggts03f7ttPzFJyFiHgqoJ • 🎨 ComfyUI: Watch Here: https://www.youtube.com/watch?v=70HTvc-4gek&list=PL8HbYkSF5f9Ggts03f7ttPzFJyFiHgqoJ • 🖥️ Automatic 1111: Watch Here: https://www.youtube.com/watch?v=70HTvc-4gek&list=PL8HbYkSF5f9Ggts03f7ttPzFJyFiHgqoJ • 📌 AI Tools: Watch Here: https://www.youtube.com/watch?v=70HTvc-4gek&list=PL8HbYkSF5f9Ggts03f7ttPzFJyFiHgqoJ 🌎Civitai: https://civitai.com/user/AIknowlege2go 🔗 Resources Mentioned: ComfyUI https://www.comfy.org/ Chapters 00:00 Intro 00:35 Install 01:37 Text 2 Video 08:57 Image 2 Video 13:42 Bonus – Longer Videos

Comments

Ishmil Waterman

does Wan2.2-T2V-A14B-GGUF go in diffusion models or check points? your folder graphic shows safetensors not gguf so im not 100% sure

Ishmil Waterman

thanks. running into 1 another error. using the above t2v workflow. KSamplerAdvanced BaseLoaderKJ._patch_modules..attention_sage() got an unexpected keyword argument 'transformer_options' When I set PATCH SAGE ATTENTION KJ to disable it runs, but obviously slowly without it.

Daniel Bostic

Amazing Workflow, i only ran into one issue, how can i fix this error for the frameinterpolation? GIMMVFI_interpolate 'CUDA_HOME' not set, unable to find cuda-toolkit installation.

AI Knowledge Central

Apparently you need to download and install the CUDA Toolkit. i would go with CUDA Toolkit 12.8 Update 1

Daniel Bostic

Yes, installing the cuda toolkit worked! The only other issue im having is the sageattention node before the loras, is it supposed to be "didabled" or on "auto" whenever i run it on auto i get an error

Daniel Bostic

the error says BaseLoaderKJ._patch_modules..attention_sage() got an unexpected keyword argument 'transformer_options'

Daniel Bostic

I fixed this issue, all i needed to do was to Update All in the Comfy Ui Manager. Thanks for your Great Video again.

Greg P

Getting this error during workflow - please advise - KSamplerAdvanced No module named 'triton'

B Buehler

All of my I2V come out very blurry. Any help appreciated. I am using Wan2.2-I2V-A14B-HighNoise-Q4_K_S and low noise gguf. Any suggestions appreciated.

AI Knowledge Central

Hi, can you check that you used the correct lighting loras for the high and the low model or that the loras are in the correct folders and can be found. There are four loras. For Image to video i2v high i2v low For Text to video t2v high t2v low

B Buehler

you got it...........looking good now........thank you

Logovore

I followed your 50% faster video and then ran the script to install everything from this video, and I'm getting this about halfway through gen: ComfyUI\venv\Lib\site-packages\sageattention\core.py", line 680, in sageattn_qk_int8_pv_fp8_cuda assert SM89_ENABLED, "SM89 kernel is not available. Make sure you GPUs with compute capability 8.9." Can you help? I have a 4090. If I bypass Sage in the workflow, I get this at the end: File "F:\AI\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI\gimmvfi\generalizable_INR\modules\softsplat.py", line 270, in cuda_launch raise RuntimeError("'CUDA_HOME' not set, unable to find cuda-toolkit installation.") RuntimeError: 'CUDA_HOME' not set, unable to find cuda-toolkit installation.

Anoobus 97

HI, great video, Im confused about the "diffusion models.safetensors" vs. ".GGUF" files, It seems the workflow needs GGUF files in addition to the green text screenshot but Im confused where to load/drop the files.

Anoobus 97

That worked perfectly, been trying to utilize the WAN 2.2 I2V workflow now but having mismatch channel errors, any idea why this might be happening? VAEDecode Given groups=1, weight of size [16, 16, 1, 1, 1], expected input[1, 48, 21, 53, 80] to have 16 channels, but got 48 channels instead

Anoobus 97

I think because Im using 5B TI2v, mismatching with the rest of the workflow?

Z A L I M wolfツYT

And showing this error (PROMPT OUTPUTS FAILED VALIDATION: LOADERGGUF) (REQUIRED INPUT IS MISSING: GGUF NAME LOADER GGUF)

AI Knowledge Central

Wan 2.2 does not need a clip vision model. The model doesn't have an image cross attention layer. If you provide clip embeds via the clip vision model in the workflow, they are simply ignored as there's nothing to process them in the model.

sachin paul

can it generate the nsfw content?

Nick Klimek

I'm having a hard time. When I load your workflow, the 2 gguf loaders in the top left, when I click the play triangle button, it says undefined, now I know you said that means they aren't in the correct folder, but I downloaded both the high and low, example "Wan2.2-T2V-A14B-LowNoise-Q5_K_M.gguf" and placed them both in the diffusion models folder. I have an nvidia 4070 graphics card, if that makes a difference. Thanks for all you do! Also, do any GGUF files go in to the folder "unet"? On the github, I believe it instructs me to put them in the 'unet' folder, but in your video it seems he doesn't talk about any files that should be in "unet'. You say to put them in "diffusion models" right?

Shelby Black

How do I keep the videos from getting darker as we extend them?

AI Knowledge Central

There is LongCat in ComfyUI for Videos that should be able to do even longer videos. I haven't tested it yet. So I can't say if its good or not.

Andy Walker

High im getting an error - KSamplerAdvanced: - Required input is missing: positive - Required input is missing: negative any help would be greatly appreciated