Qdrant
Qdrant is an open-source vector database and vector similarity search engine designed for storing, managing, and retrieving high-dimensional vector data. It supports efficient approximate nearest neighbor (ANN) search and is applicable to AI scenarios such as semantic search, recommendation systems, and image retrieval. Qdrant provides RESTful API and gRPC interfaces, supports hybrid queries with multiple filter conditions, and offers horizontal scalability for deployment in cloud-native environments.
Core Facts
Agent的记忆系统通常分为三个层次：工作记忆（当前对话上下文窗口）、短期记忆（近期交互历史）和长期记忆（通过向量数据库持久化存储）
Vector 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.
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
Timeline (last 90 days)
Dedicated vector databases (such as Pinecone, Qdrant, and Weaviate) offer more extreme performance optimization and distributed scaling for large-scale vector retrieval, while pgvector's advantage is introducing no new components, supporting transactional writes and JOIN queries, and keeping operational complexity low.
Operators of vector databases with continuous ingestion should upgrade directly to Qdrant v1.19.2 to eliminate WAL runaway and restart penalties.
Qdrant v1.19.2 added the ability to bind internal inter-node gRPC sync traffic to a dedicated network interface via PR #10853.
Qdrant v1.19.2 removed the global collection lock held during Raft snapshot creation via #10891, so concurrent reads are no longer blocked.
PR #10682 switched the sparse inverted index from full rewrites to an incremental append-based mode, and according to the original data, write amplification in streaming update scenarios was reduced by up to 80%.
Qdrant v1.19.2 jointly implemented persistent state management for proxy segments via PR #10349, #10445, #10661, and #10700, allowing the WAL to safely truncate old segments.
Qdrant v1.19.2 uses tokio::task::spawn_blocking via PR #10617 to offload batch encoding from async worker threads to the blocking thread pool, resolving Tokio runtime starvation.
In older versions of Qdrant, proxy segments did not persist intermediate state, so the WAL could not be truncated, and recovery time on node restart degraded from seconds to tens of minutes.
The upsert/delete operations generated by RAG Sync Monitor can be applied directly to vector databases such as Pinecone, Qdrant, Chroma, and pgvector.
CodebaseRAG adopts a hexagonal architecture, where query use cases depend on ports, and pgvector, Qdrant, Ollama and Anthropic sit behind the ports as adapters.
40 more timeline events
All Facts (20)
Agent的记忆系统通常分为三个层次：工作记忆（当前对话上下文窗口）、短期记忆（近期交互历史）和长期记忆（通过向量数据库持久化存储）
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%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%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%VerifiedDify's core architecture adopts a microservices design, using Docker Compose to decouple and deploy components such as the API service, Worker, Web frontend, PostgreSQL, Redis, and vector database.
80%VerifiedMainstream vector databases include Faiss, Milvus, Qdrant, and Weaviate, which can complete Approximate Nearest Neighbor (ANN) searches over millions of vectors in milliseconds.
75%VerifiedMainstream vector databases such as Pinecone, Weaviate, and Qdrant are implemented based on ANN algorithms such as HNSW or IVF-PQ.
70%VerifiedChroma and Pinecone are vector databases used to store vectorized data.
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%VerifiedFlowise supports mainstream vector databases including Pinecone, Weaviate, Chroma, and Qdrant.
65%UnverifiedFastEmbed与Qdrant出自同一团队，两者有天然的集成关系
95%UnverifiedFastEmbed是由Qdrant团队开发的Python库，专门用于在本地环境中快速生成文本和图像的向量嵌入
95%UnverifiedQdrant向量数据库在使用add方法添加文档时，如果没有手动提供嵌入向量，会自动调用FastEmbed在后台完成嵌入计算
90%Partially VerifiedAnythingLLM has built-in support for multiple vector databases including ChromaDB, Pinecone, LanceDB, and Weaviate.
80%Partially Verified常见的向量数据库包括Milvus、Qdrant、Weaviate等开源方案以及Pinecone等云服务
65%UnverifiedThe project's vector storage uses the Qdrant vector database.
60%UnverifiedIn production environments, when the data scale exceeds millions of vectors, it is generally necessary to consider vector database solutions such as Milvus and Qdrant that support distributed deployment and HNSW index optimization.
60%UnverifiedDedicated vector databases (such as Pinecone, Qdrant, and Weaviate) offer more extreme performance optimization and distributed scaling for large-scale vector retrieval, while pgvector's advantage is introducing no new components, supporting transactional writes and JOIN queries, and keeping operational complexity low.
50%