WAN 2.1 FusionX + Self Forcing LoRA are the New Best of Local Video Generation with Only 8 Steps + FLUX Upscaling Guide
Added 2025-06-19 00:32:19 +0000 UTCFusionX and Self Forcing LoRA are The BEST AI Video Model + FLUX Hyper-Realistic Upscaling (One-Click Setup!). Struggling to create high-quality AI videos and hyper-realistic images? This tutorial is your ultimate solution! I'm introducing the incredible new Wan 2.1 FusionX + Self Forcing LoRA models and a game-changing 2x latent upscaler for the FLUX model, all made incredibly simple with my custom one-click presets for SwarmUI.
Self Forcing LoRA works with same presets of FusionX
Tutorial Video : https://youtu.be/Xbn93GRQKsQ
🔗Follow below link to download the zip file that contains SwarmUI installer and AI models downloader Gradio App (FusionX Models + Self Forcing LoRA) - the one used in the tutorial ⤵️
▶️ How to install SwarmUI main tutorial : https://youtu.be/fTzlQ0tjxj0
🔗Follow below link to download the zip file that contains ComfyUI 1-click installer that has all the Flash Attention, Sage Attention, xFormers, Triton, DeepSpeed, RTX 5000 series support ⤵️
🔗 Python, Git, CUDA, C++, FFMPEG, MSVC installation tutorial - needed for ComfyUI ⤵️
🔗 SECourses Official Discord 10500+ Members ⤵️
🔗 Stable Diffusion, FLUX, Generative AI Tutorials and Resources GitHub ⤵️
🔗 SECourses Official Reddit - Stay Subscribed To Learn All The News and More ⤵️
Video Chapters
0:00 Introduction to the New FusionX Video Model & FLUX Upscaling
0:30 One-Click Presets & The SwarmUI Model Downloader Explained
1:07 Achieving Hyper-Realism with the FLUX 2x Latent Upscale Preset
1:58 How to Download & Install the SwarmUI Model Downloader
2:49 Downloading Full Models vs. Downloading Just The LoRAs
3:48 Final Setup: Updating SwarmUI & Importing The New Presets
4:32 Generating a Video: Applying the FusionX Image-to-Video Preset
5:03 Critical Step: Correcting The Model's Native Resolution Metadata
5:55 Finalizing Image-to-Video Settings (Frame Count & RIFE Interpolation)
6:49 Troubleshooting Performance: Identifying Low GPU Usage & Shared VRAM Bug
8:35 The Solution: Disabling Sage Attention for Image-to-Video Models
10:02 Final Result: Showcasing The Amazing HD Quality Animation
10:40 How to Use the FusionX Text-to-Video Model with Presets
11:49 Text-to-Video Result & Quality Comparison
12:08 How to Use the FusionX LoRA with the Base Wan 2.1 Model
13:07 FLUX Tutorial: Downloading The Required Upscaler & Face Models
13:48 Generating a High-Quality Image with The Official FLUX Preset
14:50 Using Automatic Face Segmentation & Inpainting with FLUX
16:05 The Ultimate Upgrade: Applying The FLUX 2x Latent Upscaler Preset
16:32 Final Result: Comparing Standard vs. 2x Upscaled Image Quality
16:50 Outro & Sneak Peek of The New Ultimate Video Processing App
Discover FusionX, a powerful model that generates stunning videos from text or images in as few as 8 steps. I'll guide you through the entire process, from downloading the model (or just the LoRA) with our custom high-speed downloader to applying the optimized presets. We'll even tackle a common performance bug to ensure you're getting maximum speed from your GPU.
Then, take your FLUX generations to a level you never thought possible! If you've had trouble getting realistic results, my new presets are here to help. Learn how to generate a great base image and then apply a powerful 2x latent upscale workflow that adds breathtaking detail, quality, and realism. You won't believe the before-and-after difference.
🔥 In This Tutorial, You Will Learn:
Introducing FusionX: A deep dive into the new Wan 2.1 video model.
Easy Setup: How to use the custom Model Downloader to get all necessary files (FusionX, FLUX, Upscalers, Face Models) with a single click.
Image-to-Video Mastery: A step-by-step guide to animating images with FusionX for mind-blowing results.
Text-to-Video Made Simple: How to use the text-to-video model and LoRA for amazing animations.
FLUX Hyper-Realism: Generate stunningly realistic images using the official FLUX Dev preset.
Ultimate Upscaling: Apply the 2x Latent Upscaler preset to add incredible detail and quality to your FLUX images.
Troubleshooting: How to fix common performance issues (like the shared VRAM bug) for optimal generation speed.
This guide provides everything you need to start creating professional-grade AI content today. No more guesswork, just incredible results.
Thank you for watching! If this tutorial helped you, please leave a Like, Subscribe for more advanced AI content, and share your amazing creations in the comments below!
Comments
i just tested and it downloads following models. so you must had some temporary error try again Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Queued: Wan 2.1 T2V 1.3B FP16 Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\diffusion_models\Wan2.1_1.3b_Text_to_Video.safetensors' already exists and overwrite/pre-delete not allowed. Skipping download for 'Wan 2.1 T2V 1.3B FP16'. Queued: Wan 2.1 CausVid T2V/I2V LoRA v2 14B (Rank 32) - Companion INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\Lora\Wan21_CausVid_14B_T2V_lora_rank32_v2.safetensors' already exists and overwrite/pre-delete not allowed. Skipping download for 'Wan 2.1 CausVid T2V/I2V LoRA v2 14B (Rank 32) - Companion'. Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Queued: Wan 2.1 CausVid T2V/I2V LoRA 14B (Rank 32) - Companion INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\Lora\Wan21_CausVid_14B_T2V_lora_rank32.safetensors' already exists and overwrite/pre-delete not allowed. Skipping download for 'Wan 2.1 CausVid T2V/I2V LoRA 14B (Rank 32) - Companion'. Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Queued: Phantom Wan 14B FusionX LoRA INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\Lora\Phantom_Wan_14B_FusionX_LoRA.safetensors' already exists and overwrite/pre-delete not allowed. Skipping download for 'Phantom Wan 14B FusionX LoRA'. Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Queued: Wan 2.1 I2V 14B FusionX LoRA INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\Lora\Wan2.1_Image_to_Video_14B_FusionX_LoRA.safetensors' already exists and overwrite/pre-delete not allowed. Skipping download for 'Wan 2.1 I2V 14B FusionX LoRA'. Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Queued: Wan 2.1 T2V 14B FusionX LoRA INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\Lora\Wan2.1_Text_to_Video_14B_FusionX_LoRA.safetensors' already exists and overwrite/pre-delete not allowed. Skipping download for 'Wan 2.1 T2V 14B FusionX LoRA'. Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Queued: Wan 2.1 14B Self Forcing LoRA T2V/I2V Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Queued: Wan 2.1 T2V 14B 720p GGUF Q6_K INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\Lora\Wan21_14B_Self_Forcing_LoRA_T2V_I2V.safetensors' already exists and overwrite/pre-delete not allowed. Skipping download for 'Wan 2.1 14B Self Forcing LoRA T2V/I2V'. INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\diffusion_models\Wan2.1_14b_Text_to_Video_GGUF_Q6_K.gguf' already exists and overwrite/pre-delete not allowed. Skipping download for 'Wan 2.1 T2V 14B 720p GGUF Q6_K'. Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Queued: Wan 2.1 I2V 14B 720p GGUF Q6_K Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\diffusion_models\Wan2.1_14b_Image_to_Video_720p_GGUF_Q6_K.gguf' already exists and overwrite/pre-delete not allowed. Skipping download for 'Wan 2.1 I2V 14B 720p GGUF Q6_K'. Queued: UMT5 XXL FP8 Scaled (Default for SwarmUI) Starting download: UMT5 XXL FP8 Scaled (Default for SwarmUI)... -> Downloading file 'split_files/text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors' from Comfy-Org/Wan_2.1_ComfyUI_repackaged into 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\clip' (preserving structure from filename)... Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Directory check for download: Directory check complete for 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models' (ComfyUI mode: False). Created: 0, Verified Existing: 16. Queued: CLIP Vision H (Used by Wan 2.1) Bundle 'Wan 2.1 Core Models Bundle (GGUF Q6_K + LoRA)' processed. Queued: 11, Errors: 0. umt5_xxl_fp8_e4m3fn_scaled.safetensors: 100%|██████████████████████████████████████| 6.74G/6.74G [01:06<00:00, 101MB/s] -> File downloaded to actual path: E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\clip\split_files\text_encoders\umt5_xxl_fp8_e4m3fn_scaled.safetensors -> Renaming 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\clip\split_files\text_encoders\umt5_xxl_fp8_e4m3fn_scaled.safetensors' to 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\clip\umt5_xxl_fp8_e4m3fn_scaled.safetensors'... -> Final target path E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\clip\umt5_xxl_fp8_e4m3fn_scaled.safetensors exists. Removing before rename... -> Successfully renamed to: E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\clip\umt5_xxl_fp8_e4m3fn_scaled.safetensors SUCCESS: Downloaded and processed UMT5 XXL FP8 Scaled (Default for SwarmUI) in 67.56 seconds. Final location: E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\clip\umt5_xxl_fp8_e4m3fn_scaled.safetensors INFO: Final target file 'E:\SwarmUI_Model_Downloader_v29\SwarmUI\Models\clip_vision\clip_vision_h.safetensors' already exists and overwrite/pre-delete not allowed. Skipping download for 'CLIP Vision H (Used by Wan 2.1)'.
Furkan Gözükara
2025-07-01 23:31:21 +0000 UTCty will make a tutorial hopefully with installer
Furkan Gözükara
2025-07-01 22:52:00 +0000 UTCI am not sure if you are aware of this but I stumbled into it looking for a way to make an nVidia k1 work with AI for cheap (hint it does not): https://github.com/WhatDreamsCost/Spline-Path-Control Which looks like something veeery specific and an easy way to control video motion. I wonder how easy it would be to incorporate in your custom set ups.
Christos
2025-06-26 04:35:43 +0000 UTCI downloaded the Wan 2.1 Core Models Bundle (GGUF Q6_K + LoRA). However, in the "Image To Video" Video Model drop down, I don't see what you have in the video. I just see the GGUF model. When I try to generate with it, SwarmUI gives an error: Invalid value for parameter Video Model: Invalid model value for param Video Model - 'Wan2.1_14b_Image_to_Video_720p_GGUF_Q8.gguf' - are you sure that model name is correct? Help would be greatly appreciated!
Henry
2025-06-25 08:33:19 +0000 UTCnot at the moment but i am looking for better models which may arrive sooner
Furkan Gözükara
2025-06-24 11:44:56 +0000 UTCSome of the generated videos are perfect, but several generations have lot of artifacts and shadows around the subjects. It also adds random effects like smoke. When I use the Self Forcing Lora it changes the characters face a lot. Anything I can do? (Nvidia 4090 24GB)
Ken
2025-06-24 10:56:23 +0000 UTChi SwarmUI is auto installed. but we have manual installer of comfyui too if you want and you can give its backend : https://www.patreon.com/posts/105023709
Furkan Gözükara
2025-06-22 19:36:44 +0000 UTCyes i recommend try Self Forcing LoRA + use image to video with realistic images it improves. and thanks
Furkan Gözükara
2025-06-22 19:28:07 +0000 UTCHi! I’m just starting with SwarmUI and ComfyUI, and yesterday I finally generated my first video on an RTX 5090 — emotions are off the charts! 😊 I used to run everything on Linux, but I kept getting random errors and crashes. Since I switched to SwarmUI on Windows, everything finally works smoothly. The results are amazing, but I noticed the skin in some videos looks too plastic or overly smooth. Are there any workflows or tricks to make skin look more natural and realistic in videos? Maybe combining multiple LoRAs or running several passes? Would love to hear your thoughts — thanks so much for your great work!
Oleh 23
2025-06-21 20:42:09 +0000 UTCIs there a one click install for Massed compute to install both ComyUI and Swarm with models available?
Walker4k
2025-06-21 19:28:39 +0000 UTCawesome. and thanks a lot for comment
Furkan Gözükara
2025-06-20 11:27:28 +0000 UTCour app has it but it uses last frame to extend . did you try? https://www.patreon.com/posts/123105403
Furkan Gözükara
2025-06-20 11:27:15 +0000 UTCI really appreciate the video. I'm curious if there has been any work toward wan extending an existing video. It's by far the best local gen, but the frame limitation keeps me away. I think ~15 seconds is good amount of video.
Nitty
2025-06-19 14:48:12 +0000 UTCHey, I love that I can turn on AI dubbing in Polish on your video. I really appreciate it, and it’s a great tutorial!
starysmok
2025-06-19 07:17:55 +0000 UTC
