RAG & Knowledge
zvec
Lightweight, lightning-fast in-process vector database from Alibaba that embeds in your app.
★ 16.1k starsApache-2.0
View guide →Fast, lightweight Python embeddings from Qdrant — dense, sparse, ColBERT, image and rerankers.
fastembed is an open-source rag & knowledge project by qdrant (qdrant/fastembed on GitHub). It has 3,237 GitHub stars and 266 forks, is released under the Apache-2.0 license, and was last updated on GitHub on 7 October 2026.
Decide the first concrete outcome you want from it, so you can pick the right example, template or mode instead of exploring everything at once.
Clone the repository (or install the published package) and skim its README quick start for the requirements for your platform.
Run the documented install step and set any required configuration — such as provider keys, data sources or local services — before the first run.
Execute the smallest documented example end to end, confirm the output is what you expected, then adapt it to your real data or workflow.
State-of-the-art dense vectors with automatic model download on first use.
Sparse representations for hybrid lexical/semantic retrieval.
ColBERT-style multi-vector embeddings for high-quality ranking.
Embed images, and use ColPali late-interaction multimodal models for document pages.
Built-in cross-encoder reranking to refine retrieval results.
Designed to run well on CPU, with an optional GPU package, and native Qdrant integration.
Get the source so you can inspect, configure and run it locally.
git clone https://github.com/qdrant/fastembed.git
cd fastembedInstall the standard package, or the GPU variant.
pip install fastembed
# or with GPU support: pip install fastembed-gpuCreate a TextEmbedding model and embed documents — the first call downloads and initialises the model.
from fastembed import TextEmbedding
model = TextEmbedding()
embeddings = list(model.embed(documents)) # see README quickstartIt is maintained by qdrant on GitHub (qdrant/fastembed), where the source, issues and releases live.
Yes — it is free and open source under Apache-2.0, which permits use, modification and commercial redistribution, with licence/notice retention.
No — models run locally after a one-time download, so there is no API key or per-call cost.
Compared with calling a hosted embeddings API such as OpenAI's, fastembed's distinguishing focus is: fast, lightweight Python embeddings from Qdrant — dense, sparse, ColBERT, image and rerankers. Pick it when that focus matches your need; the alternative may suit you better if its broader or different scope is the priority.
Clone https://github.com/qdrant/fastembed (or install its published package), follow the README quick start to install and configure it, then run the smallest documented example — the Run section on this page mirrors those steps.
Lightweight, lightning-fast in-process vector database from Alibaba that embeds in your app.
Meta's library for efficient similarity search and clustering of dense vectors, in C++ and Python
AI-native database for hybrid search across dense, sparse, and tensor vectors plus full text
A blazing-fast vector index in Rust, built on TurboQuant - with Python bindings
Codealive-mcp, in Python — built for context engine for large codebases, exposed throug...
Vector similarity search inside Postgres - store embeddings next to your data
The most-starred other RAG & Knowledge projects in the LibreAI catalogue — the realistic shortlist if fastembed is not the right fit.
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or…
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context…
Lightweight OCR toolkit turning PDFs and images into structured, LLM-ready data in 100+ languages
Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience
Ready-to-run templates for RAG, AI pipelines and enterprise search that stay in sync with live data
Free and open — grab the source, star the repo, and start creating.