embedding

What problem does it solve?

This Skill helps developers convert both text and images into high-quality vector embeddings for fast similarity search, clustering, and cross-modal retrieval across platforms.

Core Features & Use Cases

• Text and image vectorization: Generate embeddings for documents, prompts, and media to power search and analysis.
• Multi-model support: Switch between models like SentenceTransformer, OpenAI embeddings, and CLIP-based image embeddings.
• Use Case: Build a semantic search system that indexes product descriptions and product images into a single vector space for cross-modal retrieval.

Quick Start

Install dependencies and run a quick encoding of sample texts and an image:

• pip install sentence-transformers transformers torch
• python -c "from sentence_transformers import SentenceTransformer; m = SentenceTransformer('clip-ViT-B-32'); vecs = m.encode(['hello']); print(len(vecs[0]))"