As Gartner Predicts Vector Databases Will Be Used in 30% of Enterprise Applications by 2026, Milvus Reaches Major Milestone with 10,000+ Production Deployments
REDWOOD CITY, CA – December 18, 2025 – Zilliz recently announced that Milvus, its open-source vector database, has surpassed 40,000 GitHub stars, marking one of the fastest growth spurts in the project’s history. The milestone comes as vector databases rapidly move into production, with more than 10,000 enterprise teams using Milvus in live AI systems—including NVIDIA, Salesforce, eBay, Airbnb, and DoorDash.
“We’re grateful to everyone who supports Milvus—whether they’re running it in production, contributing code, reporting issues, or simply following the project,” said James Luan, VP of Engineering at Zilliz. “Teams depend on Milvus in demanding environments where performance, reliability, and cost all matter. As adoption grows, we’ll keep listening closely to the community and turning that trust into a platform that scales for enterprise production.”
2025 Milvus Releases Strengthen Enterprise Adoption
In 2025, Milvus delivered key architectural upgrades that directly improved performance, cost efficiency, and scalability for enterprise production workloads.
Milvus 2.5 introduced native hybrid search, unifying lexical and semantic retrieval in a single engine. This eliminated the operational overhead of maintaining separate search infrastructure, enabling teams to run keyword queries, vector searches, and metadata filters together.
Milvus 2.6 further advanced performance and cost efficiency at scale, with enhancements including:
• Tiered storage reducing costs by up to 50%
• RaBitQ 1-bit quantization cutting memory usage by 72%
• BM25 full-text engine delivering 4× faster performance than Elasticsearch
• Path Index enabling 100× faster JSON filtering
• Support for 100,000+ collections per cluster for multi-tenant deployments
Organizations running Milvus in production report sub-50ms retrieval latency across billions of vectors, while lowering infrastructure costs and improving operational efficiency.
Enterprise-Grade Managed Milvus Service: Zilliz Cloud
The rapid adoption of Milvus has also driven the growth of Zilliz Cloud, the fully managed Milvus service designed for enterprise production environments. Zilliz Cloud includes Cardinal, an advanced vector engine that delivers up to 4× higher performance than open-source Milvus while maintaining 96%+ recall with no manual tuning.
Zilliz Cloud is available across 29 global regions on major cloud providers, including AWS, Azure, and Google Cloud, and offers a 99.95% uptime SLA, SOC 2 Type II compliance, and pay-as-you-go pricing. Customers can reduce the overall total cost of ownership (TCO) by up to 50% compared to self-managed deployments.
Looking Ahead: Milvus Lake
Zilliz is developing Milvus Lake, a semantic-driven multimodal lakehouse for AI-scale data. Built on a manifest + data + index architecture, Milvus Lake will provide trillion-scale storage with predictable latency and integration with Iceberg, Hudi, Spark, and Ray for unified batch processing, semantic retrieval, and training-data preparation.
“Reaching 40,000 stars represents trust from a global developer community building production AI systems,” said Charles Xie, Founder and CEO of Zilliz. “As enterprises deploy AI at scale, they need infrastructure that delivers both performance and cost efficiency. Milvus provides that foundation.”
About Zilliz
Zilliz is the company behind Milvus, the world’s most widely adopted open-source vector database. Zilliz Cloud brings that performance to production with a fully managed, cloud-native platform built for scalable, low-latency vector search and hybrid retrieval. It supports billion-scale workloads with sub-10ms latency, auto-scaling, and optimized indexes for GenAI use cases like semantic search and RAG.
Zilliz is built to make AI not just possible—but practical. With a focus on performance and cost-efficiency, it helps engineering teams move from prototype to production without overprovisioning or complex infrastructure. Over 10,000 organizations worldwide rely on Zilliz to build intelligent applications at scale.
Headquartered in Redwood Shores, California, Zilliz is backed by leading investors including Aramco’s Prosperity 7 Ventures, Temasek’s Pavilion Capital, Hillhouse Capital, 5Y Capital, Yunqi Partners, Trustbridge Partners, and others. Learn more at Zilliz.com.
Media Contact:
Molly Chen
molly.chen@zilliz.com