ivfflat 开源项目推荐（17 个）

skyzh /vector-db-from-scratch

Learn vector search with Rust and DataFusion; the C++/BusTub track is deprecated.

datawhalechina /easy-vecdb

📚 从零开始的向量数据库原理与实践教程，在线阅读地址：https://easy-vecdb.datawhale.cc/

neurondb /neurondb

NeuronDB PostgreSQL extension: vector similarity search (HNSW, IVFFlat), embeddings, kNN, ML in SQL, and hybrid full-text + vector retrieval.

vdutts7 /bedrock-tax

How we cut our AWS Bedrock bill by 93%- replacing OpenSearch Serverless with Aurora x pgvector ($700/mo → $50/mo)

adityapathak-cubastion /Milvus-Querying

This project uses Python, Hugging Face (sentence-transformers), Milvus + Docker (container running Vector DB) to create a vector database, populate it with details of many people (names, ages, salaries, addresses and their introductions) and enable searching and querying on the database contents using Cosine-Similarity distances on IVF Flat index.

varadfromeast /pgvector_hypo

Hypothetical IVFFlat indexes for pgvector: test planner choices in EXPLAIN without building a physical index

EXARTeo /Nearest-Neighbors

A C++ implementation of efficient Nearest Neighbor search algorithms (LSH, Random Projection Hypercube, IVFFlat, and IVFPQ) optimized for high-dimensional datasets like SIFT and MNIST.

Rivestack /pgvector-bench

CLI benchmark for pgvector — measure HNSW/IVFFlat latency (p50/p95/p99), throughput, and recall@k. Single Go binary, interactive wizard, HTML report. MIT.

DONGIOVANNI7 /ANN_Vector_Search

ANN search in high dimensions!

adityapathakk /Milvus-Querying

This project uses Python, Hugging Face (sentence-transformers), Milvus + Docker (container running Vector DB) to create a vector database, populate it with details of many people (names, ages, salaries, addresses and their introductions) and enable searching and querying on the database contents using Cosine-Similarity distances on IVF Flat index.

koseemre /pgvector-benchmark

Comparison of IVFFlat and HNSW Algorithms

FlowerNeverFade /miniob-vector-search

《数据库系统设计实践》MiniOB 向量检索系统：VECTOR 内核扩展、Flask 网关与 React 可视化界面

santhibhogavalli /tata-automotive-rag

RAG pipeline for automotive product pages: Playwright ingestion, pgvector hybrid search, Ollama, FastAPI on Tata automobiles — cars, bikes, scooters and spare-parts

index-management /index-management-org

Production field guide to pgvector index management & embedding-pipeline optimization on PostgreSQL — HNSW/IVFFlat tuning, ingestion pipelines, and monitoring. Live at www.index-management.org

Amank-root /ms_marco_xi_voice_rag_pg

VoxRAG is a voice- and text-enabled Retrieval-Augmented Generation system built on MSMARCO-XI.

geozelos /rag-pgvector-tuning

Lab: measure how pgvector HNSW ef_search trades latency for recall (FastAPI + Docker, no embedding API keys). make demo → chart.

usetheoai /theodb-bench

Reproducible, falsifiable performance benchmarking for TheoDB. Immutable run bundles, versioned schemas, paired significance testing — evidence, not wins.