Elasticsearch 9.0 & 8.18: Better Binary Quantization, now GA and 5x faster than OpenSearch | ColPali, ColBERT support is included alongside JinaAI embeddings and reranking

Elasticsearch has released versions 9.0 and 8.18, which bring significant enhancements, including the introduction of Better Binary Quantization (BBQ), a high-performance alternative to traditional quantization techniques that is now generally available and offers up to 5x faster query speeds compared to OpenSearch. This release also includes enhanced support for multi-stage interaction models, out-of-the-box semantic search capabilities, and improved developer tools like a simplified API for quantized vector rescoring. The updates aim to provide faster, more efficient search results while reducing computing resources and improving the developer experience. Elasticsearch remains committed to innovation and has integrated these capabilities into Apache Lucene, maintaining its position as a leader in vector database technology. Additionally, the release offers new commands in Elasticsearch Query Language (ES|QL), such as JOIN for querying across datasets, and supports various deployment options, including Elastic Cloud and self-managed environments, while emphasizing caution in using third-party AI tools with sensitive data.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.