Description:
This workflow allows you to generate video from text.
You will find a step-by-step guide to using this workflow here: link
My other workflows for WAN: link
Resources you need:
📂Files :
For base version
T2V Model: fp16, fp8
In models/diffusion_models
For GGUF version
T2V Quant Model: Q8, Q5, Q3
In models/diffusion_models
Common files :
CLIP: umt5_xxl_fp8_e4m3fn_scaled.safetensors
in models/clip
VAE: wan_2.1_vae.safetensors
in models/vae
Speed LoRA: lightx2v_T2V_14B_cfg_step_distill_v2_lora_rank64_bf16.safetensors
in models/loras
ANY upscale model:
Realistic : RealESRGAN_x4plus.pth
Anime : RealESRGAN_x4plus_anime_6B.pth
in models/upscale_models
📦Custom Nodes :
Description
What's new? :
CFG zero star : Improved prompt adherence (exerimental, disable by default)
Bugfix :
Disable automatic prompt by default,
Adjustment on the resolution selector,
Weird color / contrast spike at 60/80 frames fixed by changing VAE decode node,
Move your notification node so that it works even with upscaling disabled.
FAQ
Comments (8)
Lora stack is still works when set to disabled.
I keep getting the warning message that "CFGZeroStarAndInit" is missing. I can't find this node, despite having installed everything to my knowledge (I also ran UmeAiRT-WAN2.1-Missing_nodes.bat).
I have the same issue
@Megasherru Version 2.0 of this workflow is coming. But for CFGZero this node is native to ComfyUI but currently in the nightly version and not stable.
@UmeAiRT how do we install cfgzero? or nightly? im on version v0.3.27
@UmeAiRT can we just disable the node? the other day it worked for me but after new updates it's missing again.
@pumaai487 Yes you can disable it, it's an optional node
@UmeAiRT Thanks ill give them a try, workflows look awesome and the guides are so organized. Wicked job dude!