Back to browse
GitHub Repository

A rust native, standalone and self-managed vector database uses HNSW data structure for fast queries and search.

10 starsRust

My attempt to make Vector search engine in Rust(350k Items, ~3.5ms Qs)

by ronakgh97·Feb 17, 2026·4 points·0 comments

AI Analysis

●●SolidWizardryNiche GemBig Brain

Self-managed vector DB in Rust with <5ms queries; Pinecone/Weaviate already own this space.

Strengths
  • Solid systems engineering: LSM Trees, HNSW graphs, SIMD optimizations, memory-mapped I/O with LRU caching (86x speedup)
  • Real benchmarks on substantial datasets (350K vectors, <1ms War & Peace, <5ms Amazon 1.4M products)
  • Crash-safe writes, concurrent access, backup/restore—production-grade reliability thinking
Weaknesses
  • Pinecone, Weaviate, Qdrant, Milvus already dominate; no clear differentiation or moat vs. competitors
  • Early-stage (caveat-laden backup, limited docs, SaaS layer 'in development'); unclear if this solves a pain Postgres pgvector users have
Category
Target Audience

Backend engineers, ML/AI researchers, developers needing self-managed vector search without vendor lock-in

Similar To

Qdrant · Weaviate · Pinecone

Post Description

I’d really appreciate technical feedback, critiques, or contributors interested

Similar Projects

Infrastructure●●Solid

Rust Database from Scratch

Unified SQL, graph, and vector engine when Neon and pgvector already split these concerns.

Bold BetBig Brain
ayoubnabil
8524d ago
Infrastructure●●Solid

Aion a Rust Database

Unifies SQL, graph, and vector data in one Rust engine, but still in early alpha.

Bold BetNiche Gem
K9987plus
601mo ago
AI/ML●●Solid

PardusDB – SQLite-like vector database in Rust

Using a single-file .pardus format with CREATE/INSERT/SELECT + SIMILARITY queries gives a very familiar developer UX for embedding storage. The combination of graph-based ANN, full transactions, thread-safety, and zero external dependencies is an uncommon and useful engineering combo for local-first AI work; it would win more attention with benchmark comparisons and richer ecosystem integrations (connectors/clients).

Niche GemWizardry
JasonHEIN
204mo ago