Skip to content

🌟 What is Spector?

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 — combining vector similarity, keyword search, and hybrid ranking in a single embeddable library with zero external dependencies. Connect any AI agent via the built-in MCP server, or embed directly in your application.

Spector is an open-source, high-performance search engine built entirely on modern Java 25. It's designed for developers who want sub-millisecond search, native AI agent integration, and zero infrastructure complexity. Drop in a JAR, write a few lines of code, and you have production-grade hybrid search with built-in agent support.

🎯 What It Does

Spector indexes documents with their vector embeddings and text content, then retrieves them using multiple strategies — directly from AI agents or your application code:

graph LR
    subgraph Clients
        MCP["🤖 AI Agent (MCP)"]
        REST["🌐 REST API"]
        SDK["📦 Java SDK"]
    end

    subgraph Search Modes
        A[Vector Search] --> D[Results]
        B[Keyword Search] --> D
        C[Hybrid Search] --> D
    end

    subgraph Engines
        A --> E[HNSW ANN]
        B --> F[BM25 Scoring]
        C --> E
        C --> F
        C --> G[RRF Fusion]
    end

    MCP --> A & B & C
    REST --> A & B & C
    SDK --> A & B & C
Mode How It Works Best For
🧠 Vector Search HNSW approximate nearest neighbor graphs Semantic similarity
📝 Keyword Search BM25 scoring with term frequency saturation Exact term matching
🧬 Hybrid Search Combines both via Reciprocal Rank Fusion Best-of-both-worlds
🤖 RAG Pipeline Ingest → chunk → embed → retrieve → context assembly LLM applications
🏛️ SpectorIndex IVF-HNSW-SVASQ adaptive hybrid index Scale + recall

💎 Key Differentiators

🤖 Agent-Native (MCP Protocol)

Includes a built-in Model Context Protocol server with 6 tools. AI agents connect directly via JSON-RPC — no Python frameworks, no network round-trips.

Feature Python Vector DB MCP Spector MCP
Search latency 2–10ms 88µs p50 (23–113× faster) †
Network overhead HTTP/gRPC round-trip Zero (in-process)
Concurrent queries Limited by Python GIL 61,000 QPS †
Dependencies Python framework stack Single JAR

† Measured. See Benchmarks.

Tip

See the MCP Server Guide to connect Claude Desktop, Cursor, or any MCP client in minutes.

📦 Pure Java, Zero Dependencies

Unlike most vector databases that rely on C++, Rust, or Python bindings, Spector is 100% Java. It uses the JDK's own Vector API for SIMD acceleration — no JNI, no native libraries, no external infrastructure.

Tip

Add the JAR to your classpath and you're done. No Docker, no clusters, no ops.

🚀 Modern JVM Technologies

Technology Purpose
Java Vector API SIMD-accelerated math (AVX2/AVX-512/NEON)
Panama FFM Zero-copy memory-mapped storage, GPU interop
Virtual Threads Millions of concurrent operations without thread pools
Structured Concurrency Safe parallel task management

⚡ Sub-Millisecond at Scale

HNSW at 100K documents (128 dimensions, top-10, M=16, efSearch=64):

Search Type Average Latency Throughput
Vector 0.13 ms 7,556 QPS
Keyword 0.98 ms 1,019 QPS
Hybrid 1.01 ms 994 QPS

SpectorIndex (IVF-HNSW-SVASQ) at 10K documents (4096-dim real Qwen3 embeddings):

Config Average Latency Throughput Recall@10
nCentroids=128, nProbe=4 0.46 ms 2,173 QPS 1.0000
nCentroids=64, nProbe=4 0.62 ms 1,601 QPS 1.0000
nCentroids=128, nProbe=16 1.26 ms 792 QPS 1.0000

Note

SpectorIndex achieves perfect recall while searching only 3.1% of the data (nProbe=4 out of 128 centroids). Ingestion is 28–160× faster than standalone HNSW. Numbers measured on 24-core x86, AVX2, Java 25, ZGC with Qwen3-embedding real vectors. For comprehensive, multi-centroid sweeps and adaptive HNSW shard promotion benchmarks, see the dedicated Large-Scale Real-Embedding Benchmarks page.

🏠 Dual Deployment Modes

Mode Description Best For
Embedded In-process library, zero network overhead Microservices, desktop apps, edge
Server REST API with CORS, auth, and metrics Teams, multi-language clients

🗜️ Advanced Quantization (SVASQ + IVF-PQ)

Spector offers two quantization paths:

  • SVASQ (Vectorized Affine Scalar Quantization): Uses the Fast Walsh-Hadamard Transform to spread variance before INT8 quantization, achieving 4× compression with near-lossless recall (~97–99.5%). Used inside SpectorIndex shards.
  • IVF-PQ (Product Quantization): Provides 32× memory compression for billion-scale datasets.

Important

SVASQ gives INT8 the precision of INT12–16 by rotating vectors before quantization. See the SVASQ Deep Dive for the full theory.

📊 How Spector Compares

Latency Comparison (100K docs, 128-dim, top-10)

Engine Language Vector Avg Vector P99
⚡ Spector Java 25 0.13 ms 0.26 ms
hnswlib C++ 0.1–0.5 ms ~1 ms
FAISS C++ 0.2–0.8 ms 1–2 ms
Lucene 9+ Java 1–5 ms 5–10 ms
Elasticsearch 8+ Java 2–10 ms 10–25 ms
Qdrant Rust 2–5 ms 10–25 ms
Milvus Go/C++ 3–10 ms 10–35 ms

Note

Spector's vector search latency is competitive with native C++ implementations (hnswlib, FAISS) for in-process workloads. Numbers for external systems are from published benchmarks and ann-benchmarks.com. Hardware and configuration differences apply — these are directional comparisons, not controlled A/B tests.

Feature Comparison

Feature Spector Elasticsearch Qdrant Milvus hnswlib
Deployment Embedded + Server Cluster only Server only Cluster only Embedded only
MCP Server ✅ Built-in (6 tools) ❌ ❌ ❌ ❌
Hybrid Search ✅ RRF built-in ✅ RRF ✅ Sparse+Dense ✅ RRF ❌
Zero Dependencies ✅ JDK only ❌ Heavy stack ❌ Tokio runtime ❌ etcd, MinIO, Pulsar ✅ Header-only
Virtual Threads ✅ Project Loom ❌ Platform threads N/A (Rust async) N/A (Go goroutines) N/A
GPU Acceleration ✅ CUDA (Panama FFM) ❌ ✅ Vulkan (indexing) ✅ CUDA (search + indexing) ❌
Quantization ✅ Scalar INT8 + IVF-PQ ✅ BBQ + Scalar + DiskBBQ (IVF) ✅ Scalar + Binary ✅ IVF-PQ + IVF-SQ ❌
Re-ranking ✅ LLM via Ollama ✅ Elastic Rerank + Inference API ✅ FastEmbed / ColBERT ✅ vLLM Ranker + Cross-encoder ❌
Distributed ✅ gRPC fan-out ✅ Built-in sharding ✅ Raft consensus ✅ gRPC + etcd ❌
SIMD Acceleration ✅ Java Vector API ✅ simdvec (Panama) ✅ Native SIMD ✅ AVX/NEON ✅ AVX/SSE

Note

This comparison reflects publicly available information as of May 2025. Feature availability may vary by version and deployment mode. All products are actively evolving.

🛠️ Use Cases

🤖 Agentic AI Memory

Connect AI agents (Claude, Cursor, custom) directly to Spector via the built-in MCP server. The agent autonomously ingests documents, searches for relevant context, and retrieves information — all with zero Python glue-code. "Point your LLM at Spector's MCP port, and it instantly has mathematically-perfect long-term memory."

🤖 Retrieval-Augmented Generation (RAG)

Ingest documents (PDF, HTML, Markdown), chunk them with token awareness, generate embeddings, and retrieve relevant context for LLM prompting — all through a single /api/v1/rag endpoint or the rag_query MCP tool.

🔍 Semantic Search Applications

Power product search, documentation search, code search, or any application where meaning matters more than exact keywords.

💡 Recommendation Systems

Use vector similarity to find items similar to what users have engaged with. Sub-millisecond latency makes real-time recommendations practical.

Combine keyword precision (finding exact product SKUs, error codes) with semantic understanding (finding conceptually related documents).

📱 Embedded Analytics

Drop Spector into existing Java applications without infrastructure changes. Perfect for desktop applications, microservices, or edge deployments.

✅ When to Choose Spector

Note

Choose Spector when:

  • You want AI agents to autonomously search your data (MCP integration)
  • You want sub-millisecond hybrid search without infrastructure complexity
  • Your stack is Java/JVM and you want native integration
  • You need an embedded search library with server-mode option
  • You want GPU acceleration without leaving the JVM
  • Zero external dependencies matters to your deployment

Warning

Consider alternatives when:

  • You need a managed cloud service with zero ops
  • Your team primarily works in Python/Rust/Go
  • You need built-in ML model serving

🚀 Next Steps