Skip to main content
We support CLIP model embeddings using the open source alternative, open-clip which supports various customizations. It is registered as open-clip and supports the following customizations:
ParameterTypeDefault ValueDescription
namestr"ViT-B-32"The name of the model.
pretrainedstr"laion2b_s34b_b79k"The name of the pretrained model to load.
devicestr"cpu"The device to run the model on. Can be "cpu" or "gpu".
batch_sizeint64The number of images to process in a batch.
normalizeboolTrueWhether to normalize the input images before feeding them to the model.
This embedding function supports ingesting images as both bytes and urls. You can query them using both test and other images.
LanceDB supports ingesting images directly from accessible links.
Now we can search using text from both the default vector column and the custom vector column Because we’re using a multimodal embedding function, we can also search using images