Got questions?
We have answers.

VectorLens is a native desktop application that gives you a visual interface for your vector databases. Connect it to your database and you immediately have a UI for browsing collections, running semantic searches, visualizing embeddings, and monitoring connection health — without writing a single line of code.

VectorLens currently supports ChromaDB, Milvus, Qdrant, and Weaviate — the four major open-source vector databases. Support for additional databases is on the roadmap.

VectorLens connects to OpenAI, Cohere, VoyageAI, Mistral, Gemini, Jina, and Ollama — covering the leading cloud providers as well as local/offline inference via Ollama.

VectorLens runs natively on macOS, Windows, and Linux.

VectorLens is a one-time purchase of $14.99 USD. There are no subscriptions, no monthly fees, and no usage-based charges.

Your $14.99 license includes all current features plus free updates for one year from your purchase date. After that year, your license continues to work — you just won't receive new updates unless you choose to renew.

Not currently. We do offer a full refund within 48 hours of purchase if the license hasn't been activated — so you can try it risk-free.

Yes. Contact us at support@vectorlens.dev to inquire about educational pricing.

We offer a full refund within 48 hours of purchase, provided the license has not been activated. Email support@vectorlens.dev with your order details.

No. VectorLens connects directly from your machine to your database — local or remote — with zero telemetry. We never see your data, your queries, or your embeddings.

Not for the core application. If you're connecting to a local database and using a local embedding model (e.g. via Ollama), VectorLens works entirely offline. An internet connection is only required when connecting to cloud-hosted databases or cloud embedding providers.

Yes. VectorLens supports multi-connection management, so you can save and switch between connections — including across different database types, local vs. remote environments, and different projects.

Type a natural-language query, select your embedding model, and VectorLens embeds it in real-time and returns the nearest-neighbour results from your collection — including distance scores. It's designed for rapid iteration when you're tuning retrieval or debugging a RAG pipeline.

Yes. You can export a collection or query results to CSV or JSON directly from VectorLens.

The dashboard surfaces key metadata for each collection at a glance: size, embedding dimensions, distance metric in use, and index type — a quick sanity-check without having to write a query.

Your license is tied to your purchase, not to a specific machine. If you need to transfer your activation — due to a hardware change, OS reinstall, or new computer — contact us at support@vectorlens.dev and we'll reactivate you. We aim to turn these around within one business day.

ChromaDB doesn't ship with a visual interface — the typical approach is querying through the Python client or CLI, which means writing code every time you want to inspect your data. VectorLens connects directly to your ChromaDB instance and lets you browse collections, view documents and metadata, and run semantic searches without writing a single line of code.

Qdrant ships with a basic built-in dashboard, and Weaviate has a limited console — but none of the major open-source vector databases offer a full-featured visual interface for day-to-day development work. VectorLens is a native desktop app that works across all four (ChromaDB, Milvus, Qdrant, and Weaviate) from a single interface.

Poor RAG retrieval usually comes down to one of a few things: embedding model mismatch, collection quality issues, or distance metric misconfiguration. VectorLens lets you run semantic searches against your actual collection with any supported embedding model, see distance scores for every result, and visualize your embedding space in 2D — which makes it much faster to spot the problem than inspecting raw query output.

VectorLens renders 2D projections of your vector space in real-time, so you can see cluster structure, identify outliers, and get an intuitive sense of how your embeddings are distributed — no external tooling or Python notebooks required.

Yes — that's the core use case for VectorLens. Connect to your database, open a collection, and you have immediate visual access to your vectors, metadata, distance scores, and collection health. No scripts, no CLI, no notebook setup.