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 Settings
- Draw Things
- DiffusionBee
Commercial authorization request for IP-Adapter-FaceID with licensed InsightFace models
Hello,
I am developing ELTAVYR, a pre-launch, self-hosted AI image-generation platform.
I understand that IP-Adapter code is Apache-2.0 and that the IP-Adapter-FaceID checkpoints are currently restricted to research use because they depend on non-commercial InsightFace pretrained models.
We are requesting a separate commercial license from InsightFace for the exact face-recognition models required by the pipeline.
If InsightFace grants us commercial authorization, could you please confirm whether the official h94/IP-Adapter-FaceID checkpoints may also be used commercially inside ELTAVYR?
We are specifically interested in:
IP-Adapter-FaceID for SDXL;
IP-Adapter-FaceID-PlusV2 for SDXL;
identity-preserving image generation inside a subscription application;
self-hosting without redistribution of the checkpoints;
commercial use of generated outputs.
Because ELTAVYR is currently pre-launch and has no revenue, we would like to request a no-cost, time-limited evaluation authorization that could convert into commercial permission upon launch.
If the existing checkpoints cannot be commercially authorized even with a commercial InsightFace license, please confirm whether we may use the Apache-2.0 training code and architecture to train entirely new FaceID adapter weights from scratch using commercially licensed models and datasets.
Please provide written clarification identifying which checkpoints, if any, may be used and whether additional authorization is required.
Thank you.