Instructions to use fal/AuraSR-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use fal/AuraSR-v2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-to-image", model="fal/AuraSR-v2")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("fal/AuraSR-v2", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Memory Issue with Aura SR Upscaler on MacBook M3
#5
by JanNafta - opened
Experiencing MPS backend out of memory error with Aura SR Upscaler on MacBook M3 (36GB RAM).
Error: "MPS allocated: 3.20 GB, other allocations: 41.59 GB, max allowed: 45.90 GB. Tried to allocate 1.12 GB on private pool."
Tried reducing image size and closing other apps.
How can I resolve this? Is using CPU an option? Any specific settings for M3 Macs?
Any solution for this?
Running into the same error on a new M4 Max Macbook (36GB)
Same problem (M4 Macbook Air).