Zilliz

Zilliz is a managed vector database service based on open-source Milvus, used for billion-scale vector search and trusted by over 1000 enterprise users. It offers a fully managed service that streamlines the process of deploying and scaling vector search applications, so you don't have to worry about complex infrastructure. Zilliz is designed for large-scale vector data, with distributed high-throughput capabilities and the ability to scale clusters up to 500 CUs, supporting more than 100 billion items.

Zilliz combines high performance and reliability at any scale, using optimizations like AUTOINDEX to find the right balance between recall and performance. It's 10x faster at retrieving vectors than Milvus with the Cardinal search engine. Zilliz also ensures high availability with 99.95% monthly uptime and adheres to SOC2 Type II and ISO27001 standards for strong data security.

Zilliz is designed to be easy to use, so you can set up large-scale vector similarity searches in minutes, even if you don't have deep experience. The service is highly scalable, supports role-based access control, and is available across multiple cloud providers, including AWS, Azure, and GCP.

Some of the key features include:

• Easy to Use: Set up large-scale vector similarity search services in minutes, even without deep experience.
• Optimized Milvus: Fully managed service built on Milvus, optimized for recall and performance.
• Blazing Fast: 10x faster vector retrieval speed than Milvus.
• Highly Scalable: Supports large-scale vector data, up to 500 CUs.
• High Availability: 99.95% monthly uptime.
• Security & Governance: Meets SOC2 Type II and ISO27001 standards with role-based access control.
• Built-in Embedding Pipelines: Converts unstructured data into searchable vector embeddings.
• Multi-Cloud: Available on AWS, Azure, and GCP across eight regions worldwide.
• AI Integrations: Integrates with leading AI models and frameworks to turn unstructured data into searchable vectors.

Zilliz supports a variety of use cases, including Retrieval Augmented Generation (RAG), recommender systems, text/semantic search, image similarity search, audio similarity search, video similarity search, question answering systems, molecular similarity search, and multimodal similarity search.

New users can sign up for a free account and use official SDKs to get started building AI applications fast. Pricing is based on a pay-as-you-go model, so it's great for users who want to scale their vector search applications without upfront costs.

Published on August 2, 2024