FastEmbed
A Python library that uses ONNX Runtime to generate dense, sparse and image embeddings and rerank search results.
About FastEmbed
Retrieval applications need embedding generation and reranking without adding a full PyTorch installation.
Who it’s for
- Developers building semantic and hybrid search
- Teams generating embeddings for document and image retrieval
When to consider it
Consider FastEmbed when you want local embedding and reranking APIs with ONNX Runtime and optional GPU support.
Tradeoffs & limitations
- Apache-2.0 licensed; model weights are downloaded and initialized locally, and GPU support uses the separate fastembed-gpu package.
SOTA overview · Documentation-based assessment · Sources & review method
Updates
No updates shared yet.
Discussion
Newest firstAsk a question or share how you use FastEmbed.
Keep it helpful. Community rules
Loading discussion…
Sources & review method
Documentation-based assessment · Oct 30, 2026 · Prepared with AI assistance; not a hands-on benchmark.
- FastEmbed official repository, supported tasks and local model setup
- Official FastEmbed documentation
- Apache-2.0 license
- Official repository commit history
- Maintainer organization and official domain
Checked by SOTA · AI-assisted documentation review. Selection advice is our assessment; verify current requirements for your deployment.
Import history & original evidence
FastEmbed official repository
A Python library that uses ONNX Runtime to generate dense, sparse and image embeddings and rerank search results.
Based on official pages and announcements checked on 2026-10-29. No hands-on test, performance benchmark or popularity ranking is claimed.