YoKONCy

YoKONCy / TriviumDB

Public

轻量级嵌入式数据库引擎，将向量检索、图谱与关系型元数据原生融合在同一个存储内核中，随写随用

database db node nosql pyo3
10
0
100% credibility
Found Mar 30, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

TriviumDB is a single-file embedded database combining vector search, graph connections, and JSON storage optimized for AI applications with bindings for multiple languages.

How It Works

1
🔍 Discover TriviumDB

You hear about a simple way to store and connect smart data like AI thoughts and relationships in one easy file.

2
📁 Create your data file

Pick a spot on your computer and make a new file where all your information lives safely.

3
➕ Add your first items

Put in numbers that represent ideas or pictures, plus notes about what they mean.

4
🔗 Link related ideas

Draw connections between items, like friends or steps in a story, to build a web of knowledge.

5
🔎 Find matches and links

Ask for similar ideas and see not just matches, but also connected ones too.

6
💾 Save and keep going

Everything stays safe in your file, ready to pick up anytime.

🎉 Your smart memory works!

Now your project remembers, connects, and finds exactly what you need effortlessly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is TriviumDB?

TriviumDB is a Rust-powered embedded database that packs vector search, graph relationships, and relational JSON metadata into a single file. You insert embeddings alongside payloads, link nodes with directed weighted edges, and query via hybrid searches mixing cosine similarity, graph expansion, and MongoDB-style filters. Python and Node.js bindings let you use it from anywhere, with support for f32, f16, or u64 vectors.

Why is it gaining traction?

It fuses vector, graph, and text search natively—no stitching separate tools like Pinecone and Neo4j. Standout hooks include Cypher-like graph queries, advanced pipelines with residual refinement and diversity reranking, plus mmap storage for fast local AI workloads. Benchmarks hit solid speeds on 10k-node graphs, all in a lightweight crate.

Who should use this?

AI engineers building RAG pipelines needing graph-aware retrieval, backend devs prototyping local knowledge graphs for LLMs, or ML teams wanting a single-file store for embeddings with relationships over bloated cloud vector DBs.

Verdict

Promising for Rust fans chasing multimodal embedded storage, but at 10 stars and 1.0% credibility, it's alpha—thin docs, basic tests, expect rough edges. Fork and contribute if hybrid vector-graph fits your stack.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.