qdrant/fastembed
A lightweight Python library for generating text embeddings using ONNX Runtime, built by Qdrant.
Not currently ranked — collecting fresh signals.
star history
FastEmbed generates state-of-the-art text embeddings without requiring GPU or large PyTorch dependencies. It uses the ONNX Runtime for fast, accurate inference and supports popular embedding models including Flag Embedding, which ranks on the MTEB leaderboard. The library provides both query and passage prefixes for retrieval tasks and integrates with vector databases like Qdrant for RAG workflows.
Frequently asked
- What is qdrant/fastembed?
- A lightweight Python library for generating text embeddings using ONNX Runtime, built by Qdrant.
- Is fastembed open source?
- Yes — qdrant/fastembed is open source, released under the Apache-2.0 license.
- What language is fastembed written in?
- qdrant/fastembed is primarily written in Python.
- How popular is fastembed?
- qdrant/fastembed has 3.1k stars on GitHub.
- Where can I find fastembed?
- qdrant/fastembed is on GitHub at https://github.com/qdrant/fastembed.