Home

⚡ Welcome to Spector

The Zero-Overhead, Agent-Ready AI Memory Backbone.

Welcome to the Spector documentation. Whether you're connecting AI agents via MCP, building RAG pipelines, or need sub-millisecond search with zero infrastructure — you're in the right place.

---

🗺️ Choose Your Path

🚀 I want to use Spector

Page	Description
[[Quick Start	Getting-Started--Quickstart]]
[[Installation	Getting-Started--Installation]]
[[MCP Server Guide	Sdk-Usage--Mcp-Server]]
[[Configuration	Configuration--Parameters]]
[[REST API Reference	Api-Reference--Rest-Endpoints]]
[[Cognitive Memory	Memory]]

🧠 I want to understand how it works

Page	Description
[[Architecture Overview	Architecture--Overview]]
[[Core Concepts	Architecture--Core-Concepts]]
[[SVASQ Quantization	Deep-Dives--Svasq-Deep-Dive]]
[[Memory Internals	Memory--Architecture]]
[[6-Phase Scoring Pipeline	Memory--Scoring-Pipeline]]
[[Real-Embedding Benchmarks	Deep-Dives--Real-Embedding-Benchmarks]]

🤝 I want to contribute

Page	Description
[[Contributing Guide	Operations--Contributing]]
[[JDK API Status	Getting-Started--Jdk-Api-Status]]
FAQ	Common questions answered

---

🔥 Key Numbers

Metric	Value
🧠 Cognitive Recall	0.13ms p50 at 1M memories
⚡ Vector Search	88µs p50 (10K docs, 128-dim)
🚀 Peak QPS	61,011 concurrent searches
🤖 MCP Tools	13 tools (6 search + 7 cognitive memory)
🗜️ Compression	4×–32× (SVASQ-8 to IVF-PQ)
✅ Test Suite	685+ tests, all passing
📦 Dependencies	Zero (JDK only)

---

💡 How It Works

graph LR
    A["🤖 AI Agent"] --> B["📡 MCP Server"]
    B --> C["⚡ SpectorEngine"]
    C --> D["🧠 Hybrid Search"]
    D --> E["🎯 RRF Fusion"]
    E --> F["🤖 LLM Re-ranking"]
    F --> G["✨ Results"]
    
    H["📄 Document"] --> I["🧩 Chunking"]
    I --> J["🧬 Embedding"]
    J --> C

Loading

Tip

New here? Start with Quick Start to build and run your first search in under 5 minutes. Want to connect an AI agent? See the MCP Server Guide.

---

🌟 Project Stats

Language	Java 25
License	Apache 2.0 · BSL 1.1 (memory module)
Modules	25 Maven modules
Dependencies	Zero (JDK only)
SIMD	AVX2 / AVX-512 / NEON
GPU	CUDA via Panama FFM
MCP	Built-in, 13 agent-ready tools
Distributed	gRPC fan-out + consistent hashing

---

Built with ⚡ by Spectrayan · GitHub · Apache 2.0 · BSL 1.1 (memory)