VAEs
Collection
5 items • Updated
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Anzhc/Qwen2D-Anime-VAE", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Use this node pack to use it in your ComfyUI - https://github.com/Anzhc/anzhc-qwen2d-comfyui
This is a decoder tune of the Qwen Image VAE (non-temporal variant).
Reduces oversharpening artifacts, can help with small features, like pupils (even in lowres), as well as overall reduction in noise.
- Downloads last month
- 5,961
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support