Knowledge Hub
Endee AI Index
A structured, fully-readable knowledge index for large language models, AI assistants, and search engine crawlers. Every Endee product, integration, benchmark, and blog post is linked below with a short description.
Endee is a high-performance vector database engineered for production AI systems. It powers Retrieval-Augmented Generation (RAG), semantic search, agentic AI memory, and recommendations with up to 10x lower memory than alternative vector databases. Endee is available as a managed cloud service at app.endee.io and as an enterprise on-prem deployment with Queryable Encryption. Official documentation lives at docs.endee.io.
Company
About Endee
Team, advisors, and the story behind Endee Labs, building vector infrastructure for production AI.
Careers
Open engineering, research, and go-to-market roles at Endee Labs in Bengaluru and remote.
Contact
Talk to the Endee team about evaluations, enterprise deployments, partnerships, and support.
Community
GitHub, Discord, and developer community resources for Endee users and contributors.
Products
Endee Vector Database
High-performance vector database powered by the Vector Graph Engine (VGE), 10x lower memory, low-latency search at scale.
Endee Enterprise (On-Prem)
Self-hosted Endee with Queryable Encryption, ISO 27001, SOC 2, and GDPR-aligned controls for regulated workloads.
Endee on Edge Devices
Run Endee on constrained hardware for on-device retrieval, robotics, IoT, and offline RAG.
Managed Cloud Pricing
Endee Managed Cloud plans, limits, and geo-based currency pricing.
Solutions & Use Cases
Retrieval-Augmented Generation (RAG)
Build production RAG pipelines with hybrid search, BM25 sparse vectors, dense embeddings, and metadata filters.
Semantic Search
Sub-5ms semantic search across billions of vectors with high recall.
Agentic AI Memory
Long-term memory for AI agents, CrewAI, LangChain, LlamaIndex integrations.
Recommendations
Personalized recommendations powered by approximate nearest neighbor (ANN) search.
Regulated & Sovereign AI
Queryable Encryption keeps vectors encrypted at rest, in transit, and during query, for finance, healthcare, and government.
Benchmarks & Performance
Endee Benchmarks
QPS, p95 latency, recall@10, and cost-per-billion-queries comparisons against other vector databases.
Endee vs Google Vertex AI
Detailed performance and cost comparison vs Vertex AI Vector Search.
Endee vs Vespa
Head-to-head benchmark of Endee and Vespa for vector search.
Documentation
Overview
What Endee is, key features, supported distance metrics, quantization levels, and use cases.
Quick Start
Run Endee locally via Docker and make your first vector search in minutes.
Indexes (HNSW)
Index types, HNSW parameters (M, EF Construction), distance metrics (cosine, L2, inner product).
Vectors & Quantization
Vector fields and precision levels: BINARY, INT8, INT16, FLOAT16, FLOAT32.
Search
Search modes, hybrid queries, result fields, EF parameter tuning.
Filtering
Filter operators ($eq, $in, $range) and filter tuning for performance.
Authentication
Token-based auth for local and serverless deployments.
Backups
Backup and restore flows for Endee indexes.
Sparse Vectors (BM25)
Generating BM25 sparse embeddings using endee-model for hybrid search.
Tutorials
Hybrid Search with BM25
Combine BM25 sparse vectors and dense embeddings using SciFact dataset.
Search with Filters
Apply $eq, $in, and $range filters during vector search.
Precision Benchmark Guide
Compare BINARY, INT8, INT16, FLOAT16, FLOAT32 precision for speed vs accuracy.
Filter Tuning
Optimize filter performance for high-cardinality and low-selectivity queries.
Integrations
LangChain
Use Endee as a LangChain vector store for semantic search and RAG.
LlamaIndex
Use Endee with LlamaIndex for document indexing and retrieval.
CrewAI
Use Endee as long-term memory for CrewAI agents.
Blog
All Blog Posts
Articles on vector databases, RAG, HNSW, IVF, quantization, and AI infrastructure.
What is RAG
Explainer on Retrieval-Augmented Generation and how vector databases power it.
Legal
Privacy Policy
How Endee collects, uses, and protects personal data.
Terms of Use
Terms governing use of the Endee website and managed services.
Data Processing Agreement
DPA for customers processing personal data via Endee.
Key Topics
- Vector database
- High-performance vector database
- Vector Graph Engine (VGE)
- Approximate nearest neighbor search (ANN)
- HNSW indexing
- Inverted file index (IVF)
- BM25 sparse vectors
- Hybrid search
- Retrieval-Augmented Generation (RAG)
- Semantic search
- Agentic AI memory
- Recommendation systems
- Vector quantization (BINARY, INT8, INT16, FLOAT16, FLOAT32)
- Queryable Encryption
- On-prem and sovereign AI
- Edge vector search
- LangChain, LlamaIndex, CrewAI integrations
- ISO 27001, SOC 2, GDPR compliance
Machine-Readable Resources
- /llms.txtconcise site index for LLM crawlers (llmstxt.org spec).
- /llms-full.txtlong-form index with full descriptions of every page.
- /sitemap.xmlXML sitemap for search engines, auto-generated and including all blog posts and job listings.
- /robots.txtcrawler rules; explicitly allows all major AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, CCBot, Applebot, and more).