Zilliz ads

Consensus, the operating system for scientific research, serves 10M+ researchers, students, and clinicians across 12,500+ universities and has answered 150M+ research questions by searching 400M+ scholarly sources. As Consensus moved to agentic research, its search stack hit its limits: ❌ Pure keyword matching missed papers that used different wording ❌ Elasticsearch required compressing vectors to control cost, which hurt retrieval quality ❌ A full re-index took 24+ hours The team needed search that understands meaning rather than just matching keywords, and stays fast and affordable at 400M+ vectors. After evaluating Elasticsearch, FAISS, Pinecone, and Zilliz Cloud, Consensus chose Zilliz Cloud to power semantic retrieval behind their agentic literature search. The results: 📈 14% higher search precision ⚡ ~45 ms P99 across 400M+ vectors 🔁 Full rebuilds: 24+ hours → ~1 hour 💰 Up to 4× lower storage costs "Zilliz Cloud gives our research agent fast, high-quality semantic retrieval, directly widening the evidence it can reach." — Christian Salem, Co-founder & CPO We're proud that Zilliz Cloud powers semantic search for Consensus's literature research agent, and we're excited to keep building together as agentic research scales. 🔗 Read the full story: https://lnkd.in/gzSRyp5i Try Zilliz Cloud now: https://lnkd.in/gvSeHQMg

🎬 𝗧𝘆𝗽𝗲 𝗮 𝗰𝗵𝗮𝗿𝗮𝗰𝘁𝗲𝗿, 𝗺𝗼𝗼𝗱, 𝗼𝗿 𝘀𝗰𝗲𝗻𝗲. 𝗙𝗶𝗻𝗱 𝘁𝗵𝗲 𝗿𝗶𝗴𝗵𝘁 𝗺𝗼𝗺𝗲𝗻𝘁 𝗳𝗿𝗼𝗺 𝗺𝗼𝗻𝘁𝗵𝘀 𝗼𝗳 𝗔𝗜 𝗰𝗿𝗲𝗮𝘁𝗶𝗼𝗻𝘀. For OpenArt AI’s 8M+ creators, generation history is more than an archive. Every past image and video can become material for the next character, scene, or story. But making that history searchable became increasingly difficult. Elasticsearch queries fanned out across a growing number of indices, pulled data from frozen storage, and reached 25 seconds at P99. Searching both the creator’s prompt and the generated asset also required a hand-built dual-kNN fusion layer. OpenArt moved the workload to Zilliz Cloud. Each creation is now represented by two signals: a prompt vector for what the creator requested and an asset vector for what the model produced. Zilliz Cloud searches both together in one weighted request. 🚀 𝗞𝗲𝘆 𝗥𝗲𝘀𝘂𝗹𝘁𝘀 ⚡ 𝗥𝗼𝘂𝗴𝗵𝗹𝘆 𝟯𝟬𝟬 𝗺𝘀 𝗮𝘁 𝗣𝟵𝟵, down from 25 seconds 💰 𝗥𝗼𝘂𝗴𝗵𝗹𝘆 𝟴𝟱% 𝗹𝗼𝘄𝗲𝗿 𝗰𝗼𝗺𝗽𝘂𝘁𝗲 𝗰𝗼𝘀𝘁 📦 𝗥𝗼𝘂𝗴𝗵𝗹𝘆 𝟰𝟱𝟲 𝗺𝗶𝗹𝗹𝗶𝗼𝗻 𝘃𝗲𝗰𝘁𝗼𝗿𝘀 migrated without re-embedding 🧩 𝗡𝗼 𝗺𝗼𝗿𝗲 𝗰𝘂𝘀𝘁𝗼𝗺 𝗳𝘂𝘀𝗶𝗼𝗻 𝗹𝗮𝘆𝗲𝗿 to maintain What was once a growing infrastructure project is now a fast, meaning-based search experience. A creator’s entire library can function as one creative memory. 🔗 Full story: https://lnkd.in/gwKDYKzC 👉 Try Zilliz Cloud Free: https://zilliz.com/cloud ——— 👉 Follow Zilliz for vector database and vector lakebase updates built for production AI. #ScaleWithZilliz #VectorDatabase #AICreators

𝟮𝗠+ 𝗱𝗲𝘃𝗶𝗰𝗲𝘀. 𝟭𝟳𝟬+ 𝗰𝗼𝘂𝗻𝘁𝗿𝗶𝗲𝘀. 𝗙𝗿𝗼𝗺 $𝟭𝗠 𝘁𝗼 $𝟭𝟬𝟬𝗠+ 𝗔𝗥𝗥 𝗶𝗻 𝗿𝗼𝘂𝗴𝗵𝗹𝘆 𝘁𝘄𝗼 𝘆𝗲𝗮𝗿𝘀. Plaud is one of the fastest-growing AI hardware companies in the world, turning everyday recordings into searchable, long-lived personal memory. Making that work at Plaud’s scale is where retrieval gets serious. We’re proud that 𝗭𝗶𝗹𝗹𝗶𝘇 𝗖𝗹𝗼𝘂𝗱 provides the retrieval foundation behind Plaud’s fast-growing AI memory. As Plaud grew from hundreds of millions to billions of chunks per region, OpenSearch could no longer keep up with the performance and cost requirements of billion-scale semantic and hybrid retrieval. Plaud needed infrastructure that could handle heavy continuous writes, low-latency reads, scalar filtering, multi-region data residency, customer-managed encryption, and zero-downtime evolution. 𝗭𝗶𝗹𝗹𝗶𝘇 𝗖𝗹𝗼𝘂𝗱 gives Plaud that foundation: ⚡ < 𝟮𝟬𝟬𝗺𝘀 average recall latency 🏗️ 𝗕𝗶𝗹𝗹𝗶𝗼𝗻-𝘀𝗰𝗮𝗹𝗲 backfills in days, while the live service stays readable 🌍 𝗜𝗻-𝗿𝗲𝗴𝗶𝗼𝗻 data residency with native multi-region deployment 🔐 𝗣𝗿𝗶𝘃𝗮𝗰𝘆-𝗳𝗶𝗿𝘀𝘁 retrieval: only semantic vectors and scalar fields go to Zilliz Cloud 🚀 𝗭𝗲𝗿𝗼-𝗱𝗼𝘄𝗻𝘁𝗶𝗺𝗲 schema changes and cluster upgrades 📈 𝗙𝗿𝗼𝗺 𝗯𝗶𝗹𝗹𝗶𝗼𝗻𝘀 𝘁𝗼 𝘁𝗲𝗻𝘀 𝗼𝗳 𝗯𝗶𝗹𝗹𝗶𝗼𝗻𝘀 of chunks without re-architecting retrieval 🛠️ 𝗟𝗼𝘄𝗲𝗿 𝗼𝗽𝗲𝗿𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗼𝘃𝗲𝗿𝗵𝗲𝗮𝗱 from a fully managed service From Charles Liu, Co-founder & CTO at Plaud: “Zilliz Cloud gives us a solid foundation we can trust for agentic memory retrieval at a massive scale.” When consumer AI moves from one-off answers to agents that remember, retrieval infrastructure becomes the foundation. We’re excited to help Plaud scale that future. 📖 Full story: https://lnkd.in/eQvhregk 🔗 Try Zilliz Cloud Free: https://zilliz.com ——— 👉 Follow Zilliz for vector database and vector lakebase updates built for production AI. #ScaleWithZilliz #VectorDatabase #AgentMemory

Congrats to the Exa team on the $𝟐𝟓𝟎𝐌 𝐫𝐚𝐢𝐬𝐞! 🚀 The AI era needs a new kind of web search — one built for agents, not just humans — and Exa is leading the way. We're proud that Zilliz Cloud helps power that vision. Exa builds web search that gives agents more comprehensive, faster, and smarter access to information on the web, while supporting different modes of AI interaction. As Exa extends this capability into entity search for companies, people, and code, the team chose Zilliz Cloud as the retrieval infrastructure powering this layer. Entity search comes with more specialized infrastructure challenges: 𝗿𝗶𝗰𝗵 𝗺𝗲𝘁𝗮𝗱𝗮𝘁𝗮 𝘀𝗰𝗵𝗲𝗺𝗮𝘀, 𝗳𝗿𝗲𝗾𝘂𝗲𝗻𝘁 𝗱𝗮𝘁𝗮 𝘂𝗽𝗱𝗮𝘁𝗲𝘀, 𝗵𝘆𝗯𝗿𝗶𝗱 𝗿𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹, 𝗮𝗻𝗱 𝗺𝗮𝗻𝗮𝗴𝗲𝗱 𝘀𝗰𝗮𝗹𝗮𝗯𝗶𝗹𝗶𝘁𝘆. After surveying all the major vector databases in the space, Zilliz Cloud checked every box. Why 𝗭𝗶𝗹𝗹𝗶𝘇 𝗖𝗹𝗼𝘂𝗱: • Native hybrid search combined dense vectors, sparse BM25 (full-text search), and RRF reranking • Metadata filtering with complex entity filters and no latency cliffs • Performance at scale delivered millisecond latency at billion-vector scale • Managed infrastructure reduced operational overhead and accelerated developer velocity What changed after adoption: • <200ms latency, down from seconds • Less effort spent on DevOps the retrieval stack, more time for product development • Zero-downtime schema evolvement, easy to experienment with new datasets • The team can ship new search verticals faster As 𝗝𝗲𝗳𝗳𝗿𝗲𝘆 𝗪𝗮𝗻𝗴, 𝗖𝗼-𝗳𝗼𝘂𝗻𝗱𝗲𝗿 𝗼𝗳 𝗘𝘅𝗮, shared: "Zilliz Cloud has been an important part of our journey to building entity search, giving us the retrieval performance and operational simplicity we need to scale our entity search product quickly and confidently." AI agents can become a fundamental interface for work, learning, and decision-making only when they can access real-world information quickly, accurately, and reliably. We’re proud that Zilliz Cloud powers Exa’s entity search layer for that future. 📖 Full story: https://lnkd.in/gPMsAD9q 🔗 Try Zilliz Cloud Free: https://lnkd.in/gMhxk7aF --- 👉 Follow 𝗭𝗶𝗹𝗹𝗶𝘇 for vector databases that run at enterprise scale. #VectorDatabase #ScaleWithZilliz #AIInfrastructure

𝟴𝟬𝟬,𝟬𝟬𝟬+ 𝘃𝗲𝗿𝗶𝗳𝗶𝗲𝗱 𝗰𝗹𝗶𝗻𝗶𝗰𝗶𝗮𝗻𝘀. 𝟭𝟬,𝟬𝟬𝟬+ 𝗵𝗼𝘀𝗽𝗶𝘁𝗮𝗹𝘀. 𝟮𝟱 𝗺𝗶𝗹𝗹𝗶𝗼𝗻 𝗰𝗹𝗶𝗻𝗶𝗰𝗮𝗹 𝗰𝗼𝗻𝘀𝘂𝗹𝘁𝗮𝘁𝗶𝗼𝗻𝘀 𝗶𝗻 𝗮 𝘀𝗶𝗻𝗴𝗹𝗲 𝗺𝗼𝗻𝘁𝗵. We're proud to power the retrieval infrastructure behind 𝗢𝗽𝗲𝗻𝗘𝘃𝗶𝗱𝗲𝗻𝗰𝗲, America's #𝟭 medical knowledge platform, used by about 40% of U.S. physicians. OpenEvidence delivers evidence-based answers from leading medical sources at the moment physicians need them. In this context, retrieval quality is not an abstract metric. Every query has to return the right evidence immidiately, because the result can 𝗱𝗶𝗿𝗲𝗰𝘁𝗹𝘆 𝘀𝘂𝗽𝗽𝗼𝗿𝘁 𝗽𝗮𝘁𝗶𝗲𝗻𝘁 𝗰𝗮𝗿𝗲. At 25 million consultations a month and growing, that need became a serious infrastructure challenge. The retrieval layer had to combine 𝗹𝗼𝘄 𝗹𝗮𝘁𝗲𝗻𝗰𝘆, 𝗵𝗶𝗴𝗵 𝗿𝗲𝗰𝗮𝗹𝗹, 𝗛𝗜𝗣𝗔𝗔 𝗰𝗼𝗺𝗽𝗹𝗶𝗮𝗻𝗰𝗲, 𝗮𝗻𝗱 𝗰𝗼𝗻𝘁𝗶𝗻𝘂𝗼𝘂𝘀 𝗮𝘃𝗮𝗶𝗹𝗮𝗯𝗶𝗹𝗶𝘁𝘆. Their previous vector search provider could no longer deliver all four. After migrating to 𝗭𝗶𝗹𝗹𝗶𝘇 𝗖𝗹𝗼𝘂𝗱, OpenEvidence built the retrieval foundation powering every clinical answer it delivers: ⚡ 𝗦𝘂𝗯-𝟭𝟬𝗺𝘀 𝗹𝗮𝘁𝗲𝗻𝗰𝘆 𝘄𝗶𝘁𝗵 𝟵𝟵%+ 𝗿𝗲𝗰𝗮𝗹𝗹, fast enough that physicians never notice the infrastructure underneath 🏗️ 𝟰,𝟬𝟬𝟬 𝗤𝗣𝗦 𝗮𝘁 𝗽𝗲𝗮𝗸 𝗹𝗼𝗮𝗱 supporting 25M+ monthly consultations across 10,000+ hospitals 🔒 𝗛𝗜𝗣𝗔𝗔 𝗰𝗼𝗺𝗽𝗹𝗶𝗮𝗻𝗰𝗲 𝗮𝘁 𝘀𝗰𝗮𝗹𝗲 via GCP Private Service Connect with a fully private data path 💰 𝗭𝗲𝗿𝗼 𝗗𝗲𝘃𝗢𝗽𝘀 𝗼𝘃𝗲𝗿𝗵𝗲𝗮𝗱, fully managed so the team ships clinical AI features instead of managing infrastructure From Jagath Kumar, Head of Performance Engineering at OpenEvidence: "We believe AI is becoming a meaningful support layer for physicians, but the experience has to feel trustworthy, reliable, and seamless. Building that kind of product requires a strong foundation behind the scenes. Zilliz Cloud has helped us create that foundation as we continue to grow and serve hundreds of thousands of clinicians." 📖 Full story: https://lnkd.in/g4TcS2WM 🔗 Try Zilliz Cloud Free: https://lnkd.in/gMhxk7aF ——— 👉 Follow Zilliz for vector databases that run at enterprise scale. #ScaleWithZilliz #VectorDatabase #HealthTech

🚀 <𝟱𝟬𝗺𝘀 𝗹𝗮𝗻𝘁𝗲𝗻𝗰𝘆, 𝟱𝟬% 𝗰𝗼𝘀𝘁 𝘀𝗮𝘃𝗶𝗻𝗴—𝟭𝟮𝟯𝗥𝗙 𝗦𝗰𝗮𝗹𝗲𝗱 𝗩𝗶𝘀𝘂𝗮𝗹 𝗦𝗲𝗮𝗿𝗰𝗵 𝘁𝗼 𝟮𝟬𝟬𝗠+ 𝗔𝘀𝘀𝗲𝘁𝘀 𝘄𝗶𝘁𝗵 𝗭𝗶𝗹𝗹𝗶𝘇 𝗖𝗹𝗼𝘂𝗱 123RF, part of the Inmagine Group, is one of the world’s largest stock content platforms, with over 200 million images, videos, and audio files. Search is central to the 123RF experience. As AI search gained importance, the team started testing embedding-based semantic search to improve relevance. Their first setup used a KNN plugin on OpenSearch, but it brought 𝗵𝗶𝗴𝗵𝗲𝗿 𝗰𝗼𝘀𝘁𝘀, 𝘂𝗻𝘀𝘁𝗮𝗯𝗹𝗲 𝗶𝗻𝗱𝗲𝘅𝗶𝗻𝗴, 𝗮𝗻𝗱 𝗶𝗻𝗰𝗼𝗻𝘀𝗶𝘀𝘁𝗲𝗻𝘁 𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲. We’re proud that Zilliz Cloud now powers 123RF’s visual search in production. Supporting both text-to-image and reverse image search, it gives the team a more stable and cost-efficient foundation for semantic retrieval at 200M+ scale. ⚡ 𝟯𝟬–𝟱𝟬𝗺𝘀 𝗾𝘂𝗲𝗿𝘆 𝗹𝗮𝘁𝗲𝗻𝗰𝘆 at hundreds of millions of vector scale, with stable production performance 💰 𝟱𝟬%+ 𝗹𝗼𝘄𝗲𝗿 𝘀𝗲𝗮𝗿𝗰𝗵 𝗶𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲 𝗰𝗼𝘀𝘁 after migrating off OpenSearch 📦 𝟮𝟬𝟬𝗠+ 𝘀𝘂𝗽𝗲𝗿 𝗵𝗶𝗴𝗵-𝗱𝗶𝗺𝗲𝗻𝘀𝗶𝗼𝗻𝗮𝗹 𝘃𝗲𝗰𝘁𝗼𝗿𝘀 across the full image library 🔄 𝗭𝗲𝗿𝗼-𝗱𝗼𝘄𝗻𝘁𝗶𝗺𝗲 𝗶𝗻𝗱𝗲𝘅𝗶𝗻𝗴 for millions to tens of millions of new rows within hours 🛠️ 𝗢𝗽𝗲𝗿𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗳𝗿𝗲𝗲𝗱𝗼𝗺 with fully managed infrastructure and less DevOps overhead From Su-Meng Yong, the lead of 123RF's engineering team: “The fully managed version (Zilliz Cloud) really saves both my team and the developers a lot of time from having to deal with a lot of problems, a lot of self-managing of the cluster. And regarding latency — we went from an initial 100 milliseconds to now sub 30 to 50 milliseconds, a roughly 50% reduction while being able to maintain production throughput.” In the future, Zilliz plans to continue supporting 123RF’s search business in video and audio search workflows, extending the platform’s search capabilities further. 📖 Full story: https://lnkd.in/d7sdzZ6p 🔗 Try Zilliz Cloud (Managed Milvus) for free: https://lnkd.in/gMhxk7aF ——— 👉 Follow Zilliz for vector databases that run at enterprise scale. #ScaleWithZilliz #VectorDatabase #RAG