Instructions to use Wan-AI/Wan2.1-T2V-14B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Wan-AI/Wan2.1-T2V-14B with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Wan-AI/Wan2.1-T2V-14B", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
It took too long time to generate video
#7
by generalchan900 - opened
With 4 x A6000 cards, I used '''torchrun --nproc_per_node=8 generate.py --task t2v-14B --size 1280*720 --ckpt_dir ./Wan2.1-T2V-14B --dit_fsdp --t5_fsdp --ulysses_size 8 --prompt "Two anthropomorphic cats in comfy boxing gear and bright gloves fight intensely on a spotlighted stage."''' to generate videos. However, it took around 45 minutes to get a 5-second video clip. How can I improve it?
Video size and steps matters when you are generating the video.
Also it maybe not utilizing all the GPUs, it can't be that slow with 4 x A6000.
the four gpus were fully used...
did you find a solution to this? I have a similar problem with Wan2.2