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? :
ComfyUi 0.3.29 positive node fix,
LoRA fix with torch compile.
FAQ
Comments (7)
Hello,
it is really strange with version 2.0 on my rtx 4070 12GB i have no OOM, even when i say 60 steps 96 frames
same settings on 2.1 and i get 99% of OOM even if i decrease a lot the frames or steps.
Any clue why ?
What i liked in 2.1 when i managed to get it working was then tensor upscale and that there are more "optimization options"
sometime i'm really lucky and it works without OOM (nothing else that comfyui is running) and i got a new seed want to try again and boom...
Even with the previous seed not possible to make it run anymore :/
going back to 2.0, no single OOM :/
Note: you need to install the custom nodes for ComfyUI-GGUF for this to work. Otherwise, no matter what you do UnetLoaderGGUFDisTorchMultiGPU and DualCLIPLoaderGGUFMultiGPU won't load.
BUG for v2.1 (2.0 Works fine)
got All your ComfyUI nodes are missing required inputs, like vae_name, expression, format, frame_rate, noise_seed, sampler_name, etc.
same issue described detailed here https://civitai.com/models/1363672
If you have installed the mixlab nodes, delete them and see if the problem persists.
On Mac M4, I've got following error at frame-interpolation.
How to fix it ?
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, mps:0 and cpu!
Very very very good 😍, I liked it a lot, it would be great to add a node to polish the prompt, such as the "TIPO" node 😗
These are excellent workflows and are now my daily workhorses. All my vids are currently being produced with these. I get excellent performance and also love the preview step to decide if i want to continue with a generation to completion.