Summary
- Support for high availability architecture based on shared storage.
- Support product quantization for dense vectors
- Support 8 bit scalar quantization for dense vectors
- Support binary dense vectors, and provide Hamming distance similarity.
- Support search results caching and pagination.
- Allow to specify comment when creating database and index.
- Support functions including: regular expression, md5, substring, lower, upper, ltrim, rtrim, trim, char_posiition, sqrt, round, ceil, floor, IN, isnan, isinf, and isfinite.
- Improve the performance of full-text phrase search by providing dynamic query pruning.
- Allow threshold to be specified to limit search results when using BM25 / L2 / IP / Cosine.
- Support Arm64 architecture.
- Improve the performance of RAG analyzer.
- Provide IK analzyer for Chinese full text search which is popular for elasticsearch users.
- Integration into RAGFlow