Instructions to use nyu-visionx/RAE-dinov2-wReg-base-ViTXL-n08-i512 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use nyu-visionx/RAE-dinov2-wReg-base-ViTXL-n08-i512 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("nyu-visionx/RAE-dinov2-wReg-base-ViTXL-n08-i512", 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
Update config for diffusers AutoencoderRAE refactor
#2
by kashif HF Staff - opened
Update config.json to match the refactored AutoencoderRAE API in diffusers:
- Added: encoder_type, encoder_hidden_size, encoder_patch_size, encoder_norm_mean, encoder_norm_std
- Removed: encoder_cls, encoder_name_or_path (no longer needed - encoder is built from config, not downloaded during init)
This allows AutoencoderRAE.from_pretrained() to work without any network calls to download HF encoder weights separately.
bytetriper changed pull request status to merged
bytetriper deleted the
refs/pr/2 ref