The Zero-Overhead, Agent-Ready AI Memory Backbone.
Legacy search engines bolted vectors onto text databases. Spector is designed from the ground up for modern AI — leveraging Java Project Panama to achieve C++ bare-metal SIMD speeds natively, with a built-in MCP server that turns any AI agent into a search-powered reasoning machine.
Spector is structured around a modular, biologically-inspired architecture designed to bridge low-level bare-metal SIMD operations with high-level agent orchestration:
- Nucleus (Foundation): Core configurations, off-heap storage layouts (Panama MemorySegment), and standard utilities.
- Memory (Cognitive Engine): The flagship vector search and cognitive memory engine containing the 4-tier cortex, Hebbian graph, and sleep consolidation pipelines.
- Synapse (Gateway & APIs): Spring Boot entry points, Armeria-based REST/gRPC gateways, and stdio/HTTP Model Context Protocol (MCP) servers.
- Cortex (UI): Three.js and Angular-powered neural dashboard for real-time visualization of memory graphs, decay, and search metrics.
For a comprehensive analysis of the system architecture, data flows, thread scheduling model, and detailed Mermaid diagrams, see the Architecture Overview Docs.
Spector is an MCP-native engine — not an afterthought adapter. The MCP server runs in-process with the search engine (zero network, zero serialization), giving agents direct SIMD-accelerated access to 21 tools across search, memory, and RAG.
| Spector (MCP-native) | Typical MCP adapter | |
|---|---|---|
| Architecture | Engine + MCP in one JVM | Python wrapper → HTTP → DB |
| Search latency | 88µs (in-process SIMD) | 5–50ms (network + serialization) |
| Memory recall | 0.13ms (fused scoring) | 50–200ms (Mem0/Letta/Zep) |
| Tools | 21 (6 engine + 15 cognitive) | 3–5 basic CRUD |
| Cognitive features | Decay, Hebbian, consolidation, valence | Key-value store |
| GC pressure | Zero (Panama off-heap) | Full GC overhead |
Spector Memory is a biologically-inspired cognitive memory engine that gives AI agents the ability to remember, forget, consolidate, and associate — with microsecond latency and zero garbage collection pressure.
| Capability | What it does |
|---|---|
| 🧠 4-Tier Cortex | Working → Episodic → Semantic → Procedural memory |
| ⚡ 0.13ms recall at 1M memories | 15× faster than the 2ms target (vs. 50–200ms for Mem0/Letta/Zep) |
| 🔗 Fused SIMD Scoring | Similarity × importance × decay in a single pass — no truncation trap |
| 🛏️ Sleep Consolidation | Hippocampus-inspired pruning and partition rebuild |
| 😱 Emotional Valence | Amygdala-driven positive/negative/neutral tagging |
| 🚫 Zero GC | 100% off-heap Panama storage (≤0.01% overhead measured) |
| Capability | Technology | Performance |
|---|---|---|
| 🤖 Agent-Native (MCP) | Model Context Protocol · 21 tools · stdio + Streamable HTTP | Claude · Cursor · autonomous agents |
| ⚡ SIMD Search | Java Vector API (AVX2/AVX-512/NEON) | 88µs p50 · 61K QPS |
| 🧊 Off-Heap Storage | Panama MemorySegment · zero-copy I/O | 0.01% GC overhead |
| 🗜️ Quantization | SVASQ-8/4 · IVF-PQ · FWHT rotation | 4–32× compression · 99.5% recall |
| 🔍 Hybrid Search | HNSW + BM25 + RRF + LLM re-ranking | Sub-ms latency |
| 🖥️ GPU Acceleration | CUDA via Panama FFM | Optional · zero-copy transfer |
| 📦 Flexible Deployment | Embedded JAR · Standalone · Distributed | Zero to cluster in one config |
🎥 Watch the Neural Graph in action →
📊 Dashboard — 12+ live cognitive panels
🌌 Graph Explorer — 3D neural galaxy
🧠 Memory Table — browse & manage memories
🔬 Memory Detail — deep cognitive inspection
Prerequisites: JDK 25+, Maven 3.9+
git clone https://github.com/spectrayan/spector.git
cd spector
mvn clean test # Build & run all 685+ tests
mvn package -pl spector-dist -am -DskipTests # Build the distribution JARStart the MCP server (for AI agents):
java --add-modules jdk.incubator.vector \
--enable-native-access=ALL-UNNAMED --enable-preview \
-jar spector-dist/target/spector.jar \
--config spector.ymlClaude Desktop config — add to claude_desktop_config.json:
{
"mcpServers": {
"spector": {
"command": "java",
"args": [
"--add-modules", "jdk.incubator.vector",
"--enable-native-access=ALL-UNNAMED",
"--enable-preview",
"-jar", "/path/to/spector-dist/target/spector.jar",
"--config", "/path/to/spector.yml"
]
}
}
}All numbers measured on Intel Core Ultra 9 285K, Java 25, AVX2 256-bit.
| Benchmark | Result | Notes |
|---|---|---|
| Vector search p50 | 88–143µs | 10K–100K docs, HNSW M=16 |
| Cognitive recall at 1M | 0.13ms p50 | 15× better than 2ms target |
| Peak QPS (16 threads) | 61,011 | Concurrent vectorSearch |
| GC overhead | 0.01% | 1 pause / 100K searches |
| vs. Python MCP servers | 23–113× faster | In-process SIMD, zero network |
| I want to... | Start here |
|---|---|
| Use Spector | Quick Start · Installation · Configuration |
| Contribute to Spector | Developer Guide · Contributing |
| Connect an AI agent | MCP Server Guide · Claude Desktop Config |
| Add cognitive memory | Memory Overview · Getting Started · Use Cases |
| Use the Java SDK | Java SDK Guide · Spring AI Integration |
| Deploy to production | Docker Deployment · Performance Tuning |
| Extend with Enterprise | Spector Enterprise — LLM providers, Cortex dashboard, data connectors |
We welcome contributions of all kinds — code, docs, tests, benchmarks, and ideas!
- 🐛 Found a bug? → Open an Issue
- 💡 Have an idea? → Start a Discussion
- 🔧 Want to contribute code? → See CONTRIBUTING.md
- 🤖 AI-assisted PRs welcome!
This repository uses a split licensing model:
spector-memory— Business Source License 1.1 (transitions to Apache 2.0 on May 27, 2030)- All other modules — Apache License 2.0
For branding and trademark guidelines, see the NOTICE file.
See SECURITY.md for our security policy and vulnerability reporting.
See ACKNOWLEDGMENTS.md for credits to the cognitive science researchers, open-source frameworks, and AI coding tools that made Spector possible.
Built with ⚡ by Spectrayan