Text Embedding Generator
Generate vector embeddings for text and compare semantic similarity.
Frequently Asked Questions
What is a text embedding?
A text embedding is a list of numbers (a vector) that represents the meaning of a piece of text. Texts with similar meaning get vectors that point in similar directions, which is what semantic search, recommendations, clustering, and retrieval-augmented generation (RAG) rely on.
How do I generate an embedding?
Pick a model, paste your text, and click Generate. The tool shows the vector dimensions and a preview, and you can copy the full vector as JSON or NumPy or download it.
Which embedding model should I use?
BGE-M3 is a good default: it handles 100+ languages and inputs up to 8,192 tokens. For English-only text, BGE Small is fastest with 384 dimensions, BGE Base balances size and quality, and BGE Large gives the best English quality. Use the same model for every text you plan to compare.
What does the cosine similarity score mean?
Cosine similarity measures how closely two vectors point in the same direction, from -1 to 1. Higher means more similar meaning. Scores depend on the model, so only compare scores produced by the same model.
Is my text stored?
Text is sent to the ToolDock API and processed by Cloudflare Workers AI through Cloudflare AI Gateway, which keeps request logs and caches responses for up to 7 days so identical requests return faster. ToolDock does not store your text in its own database. Avoid pasting secrets or personal data.
Can I use this from code?
Yes. POST /api/v1/embeddings accepts the OpenAI embeddings request format, so the OpenAI Python and JavaScript SDKs, LangChain, and LlamaIndex work by setting the base URL to the ToolDock API. OpenAI and Qwen embedding models are also available with an API key from your account.