If you're looking for a replacement for OpenSearch, Vespa is a good option. It's a unified search engine and vector database that supports vector search, lexical search and search in structured data. Vespa is geared for building production search applications at scale, with a combination of fast vector search and machine-learned models. It's good for search, recommendations and semi-structured navigation, with features including scalable and efficient machine-learned model inference and auto-elastic data management.
Another good option is Algolia, an AI-powered search infrastructure that offers fast, scalable and easy-to-use search. It combines keyword search with vector understanding and can re-rank results dynamically based on user behavior and trends. Algolia serves a variety of industries and use cases, including enterprise search, headless commerce, mobile and voice search. With flexible pricing and abundant documentation, it's a good option for developers who want to add powerful search without a lot of fuss.
If you're looking for a vector similarity search option, Qdrant is worth a look. It's an open-source vector database and search engine designed for fast and scalable searches. Qdrant is highly scalable and can be easily integrated with popular embeddings and frameworks. It's designed for ease of use and deployment, with options for local installation with Docker and cloud deployment, so it's a good option for advanced search, recommendation systems and data analysis.