Developer-first vector platform

Relevance AI's mission is to accelerate developers to solve similarity and relevance problems through data. As our first step towards helping teams, we started with the data type that all the top tech companies use - vectors, a high dimensional representation of data used to determine similarities between data.

Some important similarity and relevance based problems that can be solved with vectors:

  • Semantic & unstructured data search
  • Recommendation systems
  • Data deduplication & matching
  • Topic modelling
  • User clustering
  • Zero shot classification
  • K-nearest neighbors similarity-based regression
  • Semantic operation
  • and many more

Experimentation-first vector platform

In the vector workflow to solve search and relevance problems, we decided to focus heavily on the foundation of all good solutions - the experimentation stage. Our experimentation-first approach helps users experiment, tune and prototype various vector weightings, configurations, data structures and vector search methods to improve their vectors. For a more in-depth exploration and comparison take a look at our article on experimentation-first vector database.

Relevance AI Workflow PhasesRelevance AI Workflow Phases
Relevance AI's Workflow
  • Fully managed API, so that you don't have to manage infrastructure or DevOps.
  • Filter and traditional keyword search, combine traditional methods with vector search to maximize search relevance.
  • Hybrid vector database, flexible data structures that allows for storing of multiple vectors and metadata in 1 dataset so that you don't have to manage multiple nearest neighbor indices or 3rd party metadata store for every dataset.
  • Flexible multi-vector search, flexible search methods that allows you to easily add and weight multiple different vectors into your search query to query against different data structures such as chunk. To find the best vectors and search methods for each problem.
  • Operations beyond vector search, clustering for topic modelling, vector averaging against a category to create category vectors.
  • Visualize and interpret vectors, visualize the biases of your vectors in multi-dimensional space with our embeddings projector and vector comparator graphs.
  • Real-time vector index, no index rebuilding or constant retraining. Vectors become searchable as soon as they are inserted.
  • Production-grade, once you have finished experimenting, and decided on the vectors and vector methods for search, recommendations or more. Deploy into either our enterprise production-grade environment for low latency vector search or your own FAISS or Elasticsearch with ease.


Free for individual use. 100K free requests for commercial use.

Sign up for your free at https://cloud.relevance.ai/sdk/api, no credit card required! You can view our pricing here at https://relevance.ai/pricing.

Data Privacy Policy

You own any data you upload to Relevance AI.

Everything you upload to Relevance AI is yours, including any vectors, code, configuration, metadata, output metrics, search results, visualisations and model weights. You can choose to log, export, publish, or delete any of these. We collect aggregate statistics across our users to improve our productβ€” we might do a database query to count how many users have used a specific endpoint to help decide if we want to improve our support for that endpoint. We treat your private data, source code, or trade secrets as confidential and private, as consistent with our Terms of Service and Privacy Policy.β€Œ

You can read more about our data security policy here.


If you require any support or would love to give us feedback, please visit our support page here!

Did this page help you?