Indexify
Open-source data extraction and indexing engine for RAG applications.
About
Indexify is an open source data extraction and indexing engine from Tensorlake, built for teams assembling retrieval-augmented generation systems. It ingests unstructured content such as documents, images, and audio and runs extraction pipelines over it, producing structured data and embeddings that downstream applications can query, with a focus on keeping indexes updated continuously as new content arrives rather than through periodic batch jobs. The engine is designed to scale out on Kubernetes for production workloads, placing it in the infrastructure layer between raw data sources and the vector databases or LLM applications that consume the results. Released under the Apache 2.0 license, it targets engineers who need ongoing ingestion and extraction rather than one-off preprocessing scripts. Tensorlake, the company behind the project, has since shifted its public focus toward sandbox infrastructure for AI agents, and the original Indexify repository is no longer available at its previous address.
Should you use Indexify?
Pick it when
Consider Indexify only if you already run it or want to study a design for continuously updated extraction and indexing on Kubernetes, where new documents, images, and audio flow through pipelines instead of batch jobs.
Look elsewhere when
Skip it for new projects: Tensorlake has shifted its focus to agent sandboxes and the original repository is no longer at its old address. Pathway covers incrementally updated RAG indexes, and Unstructured handles extraction.
Alternatives to Indexify
- Pathway
Incremental live indexing from Kafka, Postgres, Google Drive, and S3 with a published Python API, but under BUSL-1.1 rather than Apache-2.0.
- Unstructured
Apache-2.0 extraction for 60+ file formats that you can schedule yourself, covering the parsing role without a cluster engine.
- R2R
Self-hostable ingestion for text, PDFs, images, and audio with hybrid search behind a REST API, a simpler stack than Kubernetes pipelines.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- RAG & Document Retrieval
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Intermediate (3/5)
- License
- Apache-2.0
- Added
- Apr 3, 2026