en.Wedoany.com Reported - On April 28, 2026, data management and analytics software provider Actian officially launched the VectorAI DB vector database at the AI Dev 26 x SF conference. The product is specifically designed for regulated, offline, and edge computing environments. According to Actian's official press release, in benchmark tests based on the VDBBench methodology, VectorAI DB achieved over 22 times the query throughput of open-source vector databases Milvus and Qdrant Local on equivalent self-hosted hardware with 10 million vectors at 768 dimensions.
Headquartered in Round Rock, Texas, Actian is the data and AI division of HCLSoftware. The release of VectorAI DB marks Actian's official entry into the vector database market. According to company CTO Emma McGrattan, current mainstream vector databases are mostly built on cloud-first architectures, assuming always-on connectivity and distributed infrastructure, which cannot meet latency and data residency requirements. VectorAI DB enables developers to bring AI to where the data resides, deploying vector search directly on-premises where data is stored, with governance capabilities and data sovereignty built into the architecture.
VectorAI DB achieved a measured query throughput of 745.2 queries per second at a scale of 10 million vectors. Its core design principle is that cloud resources are not a prerequisite for scaling; the database can run independently on embedded systems, edge devices, and on-premises servers. Supported hardware ranges from NVIDIA Jetson and Raspberry Pi to industrial edge servers and local deployment environments. Developers can build once using a unified architecture and API for multi-cloud deployment without redesigning systems as applications scale.
The product is designed for industries with the strictest data sovereignty requirements. Manufacturers can embed AI directly into machines and sensors to run vector retrieval, healthcare institutions can process patient data locally within hospital data centers, financial institutions can run fraud detection and compliance monitoring in hybrid environments, and government agencies can maintain full data control within classified environments. The database features built-in AES-256 encryption and customer-managed API keys, supporting compliance requirements such as HIPAA, GDPR, ISO 27001, and SOC 2 Type II.
In terms of development framework support, VectorAI DB is compatible with mainstream tools such as LangChain, LlamaIndex, and Hugging Face, offering Python and JavaScript SDKs. The product supports vectorized management of multimodal data including text, images, audio, video, and documents. Actian is simultaneously offering a 30-day free trial and a free community edition. CTO McGrattan also released the O'Reilly-published report "Vector Databases for Enterprise AI," systematically explaining how vector retrieval is becoming a foundational capability for enterprise AI.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
