Laion / CLIP-ViT-B-32-laion2B-s34B-b79K

Overview of CLIP-ViT-B-32-laion2B-s34B-b79K

The CLIP-ViT-B-32-laion2B-s34B-b79K is a multimodal embedding model developed by LAION. This model is trained to understand the relationships between text and images using the English subset of the Laion-5B dataset. It can be used for various computer vision and retrieval tasks, including zero-shot image classification, where it can classify images based on text descriptions without additional training, image-text retrieval for finding similar images or texts, and image segmentation to identify and segment objects within images based on semantic meanings.

How to create embeddings using CLIP-ViT-B-32-laion2B-s34B-b79K

You can use CLIP-ViT-B-32-laion2B-s34B-b79K for embedding through OpenCLIP.

Once the vector embeddings are created, they can be stored in a vector database like Zilliz Cloud (a fully managed vector database powered by Milvus and used for image search or multimodal search.

Here are four key steps:

• Sign up for a Zilliz Cloud account for free.
• Set up a serverless cluster and obtain the Public Endpoint and API Key.
• Create a vector collection and insert your vector embeddings.
• Run a semantic search on the stored embeddings.

Create embeddings via OpenCLIP and insert them in Zilliz Cloud for multimodal search

A step-by-step guide: coming soon.

Further reading

• How Multimodal Retrieval Transforms Image Search
• Choosing the Right Embedding Model for Your Data
• Evaluating Your Embedding Model
• Training Your Own Text Embedding Model
• A Beginner's Guide to Website Chunking and Embedding for Your RAG Applications
• What is RAG?