Instructions to use h94/IP-Adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use h94/IP-Adapter with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("h94/IP-Adapter", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
I think these 2 file names are mixed ip-adapter-plus-face_sdxl_vit-h.safetensors - ip-adapter-plus_sdxl_vit-h.safetensors
#11
by MonsterMMORPG - opened
Can you verify?
Because face model behaves like ip-adapter-plus_sdxl_vit
And ip-adapter-plus_sdxl_vit behaves like face
@MonsterMMORPG do you also tested the .bin file?
i just compared for you
input img
face bin
face safetensors
ip adapter bin
ip adapter safetensors
config
hi, it is right.
ip-adapter-plus capture information about the entire imageip-adapter-plus-face only capture information about the face of the image (you can use a cropped face, will get better results)





