Milvus
Milvus is an open-source vector database system designed for the storage and retrieval of large-scale, high-dimensional vector data. Its core capabilities include efficient approximate nearest neighbor (ANN) similarity search, support for multiple index types such as HNSW and IVF, and distributed processing of massive vector datasets. It is commonly used in AI application scenarios including image recognition, natural language processing, and recommendation systems.
Core Facts
RAG（检索增强生成）技术架构将文档切片、向量化、存入向量数据库，查询时先检索相关片段再交给模型生成回答
The RAG workflow is divided into offline and online phases: offline, documents are chunked and converted into vectors stored in a vector database; online, questions are converted into vectors and the most similar text segments are retrieved via ANN algorithms and concatenated into the context.
Chroma is suitable for lightweight prototype development and local experiments, while Milvus and Pinecone target production-grade large-scale scenarios, supporting distributed deployment and efficient retrieval of hundreds of millions of vectors.
Timeline (last 90 days)
Vector databases (such as FAISS, Milvus and Chroma) are responsible for storing and efficiently retrieving vectors.
Milvus and Pinecone target production environments and support distribution and persistence.
The core capability of vector databases is approximate nearest neighbor search (ANN); Milvus supports billion-scale vectors, and PGVector suits teams already using PostgreSQL.
For beginners, starting with Redis or PostgreSQL for vector storage has a lower cost; enterprise-level scenarios are more inclined toward dedicated vector databases such as Milvus for better retrieval performance and scalability.
Boge calls RAG "a relatively common and low-cost solution in enterprises," and the course will lead learners through a complete implementation of the RAG workflow in Java code, involving vector databases such as ChromaDB and Milvus.
Using a vector database like Milvus can achieve semantic matching, whereas storing information in MySQL or Redis cannot be recalled directly through semantic means.
Two default practices for long-term memory: extraction should be placed asynchronously outside the reply path (because the facts of this conversation are already in this session's history); take a snapshot of memory at the start of each conversation (the memory block is placed before the history, and updating it midway would break the cache).
The hybrid architecture flow: user query → intent classification/regex gateway determines whether it is relational or semantic → graph engine (Neo4j/Memgraph) executes precompiled Cypher templates, returning exact 2-hop entity relations in 15 milliseconds → vector engine (Qdrant/Milvus) retrieves the top-3 target context text chunks in 20 milliseconds → a single frontier-model synthesis (at most 2,000 tokens).
Common vector database products include Pinecone, Weaviate, Qdrant, Milvus, and pgvector, which typically support metadata filtering (such as filtering by tenant ID or model version).
The EduRAG project adopts a dual-module architecture: the RAG core module (including the three processes of retrieval, augmentation, and generation) and the FAQ module (which reduces response latency and Token consumption by caching high-frequency question-answer pairs).
25 more timeline events
All Facts (20)
RAG（检索增强生成）技术架构将文档切片、向量化、存入向量数据库，查询时先检索相关片段再交给模型生成回答
90%VerifiedThe RAG workflow is divided into offline and online phases: offline, documents are chunked and converted into vectors stored in a vector database; online, questions are converted into vectors and the most similar text segments are retrieved via ANN algorithms and concatenated into the context.
80%VerifiedChroma is suitable for lightweight prototype development and local experiments, while Milvus and Pinecone target production-grade large-scale scenarios, supporting distributed deployment and efficient retrieval of hundreds of millions of vectors.
80%VerifiedVector databases (such as ChromaDB, Milvus, and Pinecone) use indexing algorithms such as HNSW and IVF to achieve millisecond-level retrieval among millions or even hundreds of millions of vectors.
80%Verified常见的向量数据库包括ChromaDB、Milvus、Pinecone、Weaviate，它们针对高维向量的近似最近邻（ANN）搜索进行了专门优化
80%VerifiedMilvus, Pinecone, and Chroma are vector databases used for storing and retrieving text vectors and are core components of the RAG architecture.
80%VerifiedVector databases (such as Pinecone, Weaviate, Milvus, etc.) convert text into high-dimensional vectors via embedding models and use cosine similarity or Euclidean distance for approximate nearest neighbor (ANN) search.
80%VerifiedMilvus is an open-source vector database that supports billion-scale vector retrieval and is suitable for production environments.
80%Verified长期记忆Agent通常基于RAG（检索增强生成）技术构建，通过向量数据库将信息转化为高维向量并持久化存储
80%VerifiedLong-term memory is typically implemented with the help of vector databases (such as Chroma, Pinecone, Weaviate, Milvus, etc.) to enable semantic retrieval.
80%Verified向量数据库的核心操作是近似最近邻搜索（ANN），通过HNSW、IVF等索引算法将相似度计算加速到毫秒级响应
80%VerifiedIn RAG implementations, text is converted into high-dimensional vectors by embedding models and stored in vector databases such as Pinecone, Weaviate, and Milvus; at query time, the closest segments are found via cosine similarity.
75%Verified典型的RAG流程包括文档切片、向量化嵌入、存入向量数据库、语义检索、上下文拼接和最终生成六个步骤
75%VerifiedMainstream vector databases include Faiss, Milvus, Qdrant, and Weaviate, which can complete Approximate Nearest Neighbor (ANN) searches over millions of vectors in milliseconds.
75%VerifiedThe engineering challenges of RAG are concentrated in chunk splitting strategies, embedding model selection, retrieval accuracy tuning (dense + sparse hybrid retrieval and reranking), and multimodal content parsing.
65%VerifiedDuring the RAG indexing phase, documents are chunked and converted into vectors via an embedding model, then stored in vector databases such as Faiss, Pinecone, and Weaviate.
65%VerifiedIn scientific research scenarios, RAG is typically used in combination with vector databases such as Pinecone and Weaviate.
65%VerifiedVector databases (such as Milvus, Chroma, and Weaviate) use approximate nearest neighbor algorithms to find content that is similar in meaning rather than identical in wording.
65%Verified向量存储常用专用向量数据库如Pinecone、Chroma、Weaviate，支持近似最近邻检索
65%VerifiedRAG技术栈由向量数据库（如Pinecone、Weaviate、Chroma）、嵌入模型、重排序等技术构成
65%