hi everyone!

Hello everyone! ?

Excited to publish my first article on the platform and excited to be a part of this great developer community! ?

---

? The Node.js vector database based on FAISS HNSW is here!

I’ve been developing a high-performance FAISS-based Node.js vector database and I’m excited to finally share it with you all! ?

? eada-cpu is optimized for the FAISS HNSW (Hierarchical Navigable Small World) index , allowing efficient KNN searches directly in Node.js , no Python dependency required.

---

? Benchmark results

**指标**	**数值**
**向量维度**	128
**向量数量**	7,000,000
**KNN搜索时间**	4.05 ms ?
**与FAISS-Python相比的性能**	快10% - 15%
**数据集大小**	~5GB
**索引时间**	1小时36分钟

? **Fully optimized for CPU! **

This benchmark runs entirely on the CPU, making it ideal for standard servers that don’t require GPU acceleration.

This enables LLM RAG applications, recommendation engines and vector searches to run efficiently and cost-effectively in Node.js.

---

? Main features

✅ Support FAISS HNSW → High-speed, accurate KNN search

✅ Pure Node.js → No Python dependency required

✅ Compatible with Windows/Linux/macOS (Intel and ARM64)

✅ Build system based on CMake for easy cross-platform support

✅ N-API and pre-built support → Install easily using the following command:

<code class="language-bash">npm i eada-cpu</code>

The above is the detailed content of hi everyone!. For more information, please follow other related articles on the PHP Chinese website!