Milvus(@milvusio)

Milvus 3.0 further expands what its retrieval engine can handle, with stronger support for in-engine...

8.5内容质量
Milvus 3.0 further expands what its retrieval engine can handle, with stronger support for in-engine...

TL;DR · AI 摘要

Milvus 3.0通过in-engine排名、聚合和多向量搜索增强检索能力，减少生产环境中的自定义逻辑需求。

核心要点

  • Milvus 3.0支持ORDER BY对ANN候选结果进行排序，减少应用层逻辑
  • StructArray结构允许单实体包含多向量列表，适配文档块和视频帧等场景
  • 稀疏检索优化采用压缩BM25索引和SINDI算法，提升检索效率

结构提纲

按章节快速跳转。

  1. §引言

    介绍Milvus 3.0在检索引擎能力上的扩展方向

  2. ·核心机制

    解析in-engine ranking和aggregation的技术实现

  3. ›多向量处理

    StructArray结构如何支持复杂实体数据存储

  4. ·稀疏检索优化

    BM25索引压缩与SINDI算法的协同作用

  5. ›生产价值

    减少过获取和自定义逻辑的工程实践收益

思维导图

用一张图看清主题之间的关系。

查看大纲文本（无障碍 / 无 JS 友好）
  • Milvus 3.0新特性
    • in-engine处理
      • 排名排序(ORDER BY)
      • 聚合分析
    • 多向量支持
      • StructArray结构
      • ColBERT向量处理
    • 稀疏检索优化
      • 压缩BM25索引
      • SINDI算法

金句 / Highlights

值得收藏与分享的关键句。

#Milvus#向量搜索#检索引擎#多向量处理#稀疏检索
打开原文

Milvus on X: "Milvus 3.0 further expands what its retrieval engine can handle, with stronger support for in-engine ranking, aggregation, multi-vector search, and sparse retrieval. In production, vector search rarely ends with top-k ANN results. Beyond retrieving candidates, applications often https://t.co/n6yQ4Y24Ak" / X

Milvus

@milvusio

Milvus 3.0 further expands what its retrieval engine can handle, with stronger support for in-engine ranking, aggregation, multi-vector search, and sparse retrieval. In production, vector search rarely ends with top-k ANN results. Beyond retrieving candidates, applications often need to sort by freshness, price, rating, inventory, or timestamp; group results by category; return facets; and work with entities that contain many vectors, such as document chunks, video frames, product images, or ColBERT token vectors. Milvus has long supported dense and sparse vector search, hybrid retrieval, and filtering. Milvus 3.0 builds on that foundation by bringing more ranking, grouping, and result-processing work closer to the retrieval path. ORDER BY sorts filtered query results by scalar fields and can also sort ANN candidates within Milvus. Query-side aggregation supports count, sum, avg, min, and max, grouped by scalar fields. Search aggregation adds faceted search over retrieved hits, returning bucket counts, aggregate statistics, and top-N sample hits per bucket. Since it operates on ANN results, facet counts are approximate; use query-side aggregation when exact counts matter. StructArray lets a single entity contain a variable-length list of structured elements and vectors while remaining one row. It fits workloads such as document chunks, video frames, product images, and late-interaction models like ColBERT. Sparse retrieval is also enhanced with compressed BM25 indexes and SINDI, an algorithm designed for learned sparse embeddings such as SPLADE. For retrieval teams, the practical benefit is less over-fetching and less custom application-side logic for ranking, grouping, faceting, multi-vector scoring, and sparse search. More in the Milvus 3.0 launch blog:

milvus.io/blog/announcin…

3:31 PM · Aug 10, 2026

443

Views

1

10

2