Instructions to use h94/IP-Adapter-FaceID with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use h94/IP-Adapter-FaceID 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-FaceID", 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
Crashing when using CN model + Lora (SD XL)
#70
by Amelina - opened
I'm unable to run ControlNet model + Lora while using SD XL checkpoints in A1111 on Google Colab.
When I run ip-adapter_face_id with ip-adapter-faceid-plusv2_sdxl it works fine.
When I run sdxl Lora separately it works. (The lora alone obviously gives no meaningful results, but it passes and changes picture somehow). Actually I can't run any version of lora from this repo alongside with plus v2 sdxl model.
But when I run both of them, A1111 crashes. I don't see any errors in the console, I just see that VRAM consumption jumps to the maximum right before a crash. Generations don't even pass the first step.
Tested with different checkpoints. SD15 versions work fine.