Hardware-Accelerated Architecture Brings Enterprise RAG to New Scale and Efficiency
Redwood City, CA, November 18, 2025 – Zilliz, the company behind Milvus, the world’s most advanced open-source vector database, today announced a strategic collaboration with Pliops to revolutionize large-scale Retrieval-Augmented Generation (RAG) deployments. By integrating Pliops’ LightningAI architecture with Milvus, enterprises can now achieve multi-billion-scale vector search at storage-level costs, breaking through traditional memory constraints that have limited GenAI applications.
As enterprises scale their AI initiatives, vector databases must handle billions of embeddings while maintaining performance and managing costs. This collaboration addresses these challenges head-on by combining Milvus’s industry-leading vector search capabilities with Pliops’ hardware-accelerated storage technology, enabling efficient large-scale context retrieval and inference for enterprise AI workloads.
Advancing Milvus for Next-Generation AI Infrastructure
Milvus powers mission-critical AI applications across various industries, including semantic search and recommendation systems, as well as agentic AI and RAG. Its distributed, cloud-native architecture is designed for massive scale, and this collaboration takes that capability even further.
Through this partnership, Zilliz will introduce several key enhancements to Milvus in collaboration with Pliops:
- Storage APIs and Tiering: Support for multi-tier storage to optimize cost and performance.
- KV Mapping: Adding a key-value abstraction layer on top of file offsets for efficient caching and retrieval.
- Dual-Tier Architecture:
- Flash Tier (Hot): Ultra-low latency and high-density access.
- S3 Tier (Cold): Reliable, globally distributed backup for cost efficiency.
These enhancements allow enterprises to significantly expand context windows, improve inference efficiency, and reduce infrastructure overhead—while maintaining the performance and reliability that Milvus users expect in production environments.
“This collaboration represents a significant leap forward in making enterprise-scale vector search economically viable,” said Charles Xie, Founder and CEO of Zilliz. “Pliops’ LightningAI introduces a breakthrough approach to scaling GenAI inference, and integrating it with Milvus unlocks truly massive context retrieval at a fraction of the traditional cost. As the creators of Milvus, we’re committed to advancing what’s possible in vector search. This collaboration gives enterprises a clear path to run larger models, access more knowledge, and deliver faster AI experiences—all without the memory limitations that have constrained GenAI until now.”
Ido Bukspan, CEO of Pliops, added, “LightningAI is designed to make AI inference scalable and affordable. Partnering with Zilliz brings the best of storage and retrieval intelligence together.”
Availability and Resources
The technical architecture and implementation details, including the Milvus RFC for Near Compute Storage, are available to the open-source community on GitHub. Organizations interested in deploying billion-scale vector search can learn more at:
- Zilliz: https://zilliz.com
- Milvus: https://milvus.io
- Pliops LightningAI: https://pliops.com
About Zilliz
Zilliz develops next-generation vector database technologies that help organizations extract value from unstructured data and accelerate AI application development. The company delivers a fully managed, multi-cloud service, Zilliz Cloud, powered by the open-source Milvus project, with deployments across major cloud platforms in more than 20 countries. Headquartered in Redwood Shores, California, Zilliz is backed by leading investors, including Aramco’s Prosperity7 Ventures, Temasek’s Pavilion Capital, Hillhouse Capital, 5Y Capital, Yunqi Partners, and Trustbridge Partners. For more information, visit zilliz.com.
Media Contact:
Molly Chen
molly.chen@zilliz.com
