Instructions to use zer0int/CLIP-SAE-ViT-L-14 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use zer0int/CLIP-SAE-ViT-L-14 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("zero-shot-image-classification", model="zer0int/CLIP-SAE-ViT-L-14") pipe( "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png", candidate_labels=["animals", "humans", "landscape"], )# Load model directly from transformers import AutoProcessor, AutoModelForZeroShotImageClassification processor = AutoProcessor.from_pretrained("zer0int/CLIP-SAE-ViT-L-14") model = AutoModelForZeroShotImageClassification.from_pretrained("zer0int/CLIP-SAE-ViT-L-14") - Notebooks
- Google Colab
- Kaggle
is this or the my-gmp one better for SD 3.5 large?
#1
by mimizukari - opened
(assuming they're compatible since all notes always mention flux ;_;)
I don't know, I haven't tried SD-3.5 as I am not much generating images, but generating models (Text Encoders) these days. Also, it may depend on the image you're making. I suggest just trying it and maybe posting your results, so somebody else with the same question will certainly be grateful for that. =)
I think out of all it seems like the gmp text one works the best for sd3.5l, improving quality & fixing the text projection error.