MongoDB.local San Francisco 2026: Ship Production AI, Faster

Summary

MongoDB announced several features aimed at streamlining AI application development across its database and search ecosystem. The Voyage 4 embedding model family is now generally available, introducing cross-model compatibility within a shared embedding space alongside the video-capable voyage-multimodal-3.5 model. For in-database operations, MongoDB Community Edition introduced Automated Embedding in public preview to eliminate external synchronization pipelines for vector generation. Vector Search gained Lexical Prefilters, replacing the knnBeta operator with a dedicated vectorSearch operator and vector data type inside $search aggregations. Additionally, MongoDB made its intelligent assistant generally available in Compass and released the source code for its mongot search engine under the SSPL.

Context

Building production AI applications introduces operational overhead around managing separate vector infrastructure, maintaining conversational context, syncing data across external systems, and applying precise lexical filters alongside vector search.

Approach / What changed

MongoDB introduced the Voyage 4 embedding family and voyage-multimodal-3.5, launched Automated Embedding in public preview for MongoDB Community Edition, introduced Lexical Prefilters using a vectorSearch operator in $search, made its intelligent assistant generally available in Compass and Atlas Data Explorer, and released the mongot search engine source code under the SSPL.

Takeaways

• The Voyage 4 model family shares an embedding space for cross-model compatibility and includes voyage-4-nano on Hugging Face, alongside voyage-multimodal-3.5 with native video support.
• Lexical Prefilters for Vector Search replace the knnBeta operator by introducing a vector data type in $search index definitions and a vectorSearch operator in $search aggregations.
• Mongot, the standalone engine powering MongoDB Search and Vector Search alongside the core mongod process, is now publicly available under the SSPL in preview.

Related reading