Elasticsearch kept every vector in RAM—cost-prohibitive at scale. One shared Milvus platform now runs 100+ workloads on 10B+ vectors.
One of Asia's largest technology companies, operating a super-app covering food delivery, in-store dining, hotel and travel booking, groceries, and instant retail across a vast network of local merchants and couriers.
As AI adoption expanded across business units, each new workload brought different requirements for vector types, index algorithms, and dataset...
“Meituan is exactly the kind of engineering-driven company that pushes open source forward: they run Milvus at a scale most teams never reach, and they give back what they learn. We're proud that Milvus is the vector foundation beneath Meituan's AI workloads, and we're excited to keep building alongside their team and the community.”
Local on-demand service platform for food delivery, e-commerce, and lifestyle services.
Vector database platform for building and scaling AI applications.
Meituan's Shared AI platform is part of this use case:
Related implementations across industries and use cases
Teachers spent 20 minutes grading one test. A vector engine now scores handwritten answers instantly, referencing 1 billion+ items.
General-purpose LLMs scored below 52% on multilingual ecommerce tasks. Shopee's own 245B-parameter Compass now handles 340B tokens monthly.
Keyword search failed on complex vehicle queries. Now, an AI pipeline rewrites messy questions to power shopping and service assistants.
General-purpose LLMs scored below 52% on multilingual ecommerce tasks. Shopee's own 245B-parameter Compass now handles 340B tokens monthly.
A phone launch, a breaking story, a Vietnamese phrase — keyword search came up short. Semantic retrieval now covers 92% of AI Search.
16.5 hours to a first reply—on 9,500 inquiries a week. Launched the day before Black Friday, AI agents now close 42% without a human.
Workflows collapsed when staff changed roles. A meta-agent now turns plain-language descriptions into live agents—no AI team needed.
Scaling AI access across a 100,000-employee workforce risked system overloads. An auto-scaling platform launched with zero downtime.
Energy market research that took weeks now closes same-day. Staff freed from chasing information to make the calls that matter.
Elasticsearch kept every vector in RAM—cost-prohibitive at scale. One shared Milvus platform now runs 100+ workloads on 10B+ vectors.
One of Asia's largest technology companies, operating a super-app covering food delivery, in-store dining, hotel and travel booking, groceries, and instant retail across a vast network of local merchants and couriers.
As AI adoption expanded across business units, each new workload brought different requirements for vector types, index algorithms, and dataset...
“Meituan is exactly the kind of engineering-driven company that pushes open source forward: they run Milvus at a scale most teams never reach, and they give back what they learn. We're proud that Milvus is the vector foundation beneath Meituan's AI workloads, and we're excited to keep building alongside their team and the community.”
Local on-demand service platform for food delivery, e-commerce, and lifestyle services.
Vector database platform for building and scaling AI applications.
Meituan's Shared AI platform is part of this use case:
Related implementations across industries and use cases
Teachers spent 20 minutes grading one test. A vector engine now scores handwritten answers instantly, referencing 1 billion+ items.
General-purpose LLMs scored below 52% on multilingual ecommerce tasks. Shopee's own 245B-parameter Compass now handles 340B tokens monthly.
Keyword search failed on complex vehicle queries. Now, an AI pipeline rewrites messy questions to power shopping and service assistants.
General-purpose LLMs scored below 52% on multilingual ecommerce tasks. Shopee's own 245B-parameter Compass now handles 340B tokens monthly.
A phone launch, a breaking story, a Vietnamese phrase — keyword search came up short. Semantic retrieval now covers 92% of AI Search.
16.5 hours to a first reply—on 9,500 inquiries a week. Launched the day before Black Friday, AI agents now close 42% without a human.
Workflows collapsed when staff changed roles. A meta-agent now turns plain-language descriptions into live agents—no AI team needed.
Scaling AI access across a 100,000-employee workforce risked system overloads. An auto-scaling platform launched with zero downtime.
Energy market research that took weeks now closes same-day. Staff freed from chasing information to make the calls that matter.