Chroma

It is an open-source search database that provides vector, sparse vector, full-text, regex and metadata search for AI applications that need fast retrieval over billions of multi-tenant indexes. It sells to software developers, engineering teams and data teams in b2b companies from startups to enterprises building AI applications. It is differentiated as built on object storage with automatic query-aware data tiering and caching for serverless auto-scaling with no manual tuning, offered as Apache 2.0 open source with a hosted cloud and local deployment option.

Key features

• Sparse vector search (BM25, SPLADE)
• Vector search for semantic similarity
• Full-text search with trigram
• Regex search
• Metadata filtering and faceted search
• Dataset versioning and forking
• A/B testing and roll-outs
• CLI tools for development
• Serverless auto-scaling infrastructure
• Automatic query-aware data tiering