Four-layer retrieval fusion powered by ONNX Runtime. No PyTorch. Sub-200ms warm search. Incremental graph updates. MCP-ready.
Most RAG solutions either drag in 2GB+ of PyTorch or limit you to single-layer vector search. VelociRAG gives you four retrieval methods — vector similarity, BM25 keyword matching, knowledge graph traversal, and metadata filtering — fused through reciprocal rank fusion with cross-encoder reranking. All running on ONNX Runtime, no GPU, no API keys. Comes with an MCP server for agent integration, a Unix socket daemon for warm queries, and a CLI that just works.
Lightning-fast RAG for AI agents. ONNX-powered, 4-layer fusion, MCP server. No PyTorch.
2026-10-27
Sterne
12
Tools
—
Uptime
—
🚀 Quick Start
MCP Server (Claude, Cursor, Windsurf)
pip install "velocirag[mcp]"
velocirag index ./my-docs
velocirag mcp
Claude Code — add to .mcp.json in your project root:
velocirag serve --db ./my-data # start daemon (background)
velocirag search "query" # auto-routes through daemon
velocirag status # check daemon health
velocirag stop # stop daemon
The daemon keeps the ONNX model + FAISS index warm over a Unix socket. First query loads the engine (~1s), subsequent queries return in ~180ms with full 4-layer fusion.
Cross-encoder reranking — TinyBERT reranker via ONNX Runtime — included in base install, no PyTorch needed. Downloads ~17MB model on first use
Incremental graph updates — file-centric provenance tracking detects what changed and only rebuilds affected nodes/edges. Cascading deletes maintain consistency across all stores (vector, graph, metadata). Multi-source support with isolated provenance per source
MCP server — Five tools (search, index, add_document, health, list_sources) for Claude, Cursor, Windsurf
Search daemon — Unix socket server keeps ONNX model + FAISS index warm between queries
Knowledge graph — Analyzers build entity, temporal, topic, and explicit-link edges from markdown. Optional GLiNER NER. 418 files in 2.1s
Reciprocal Rank Fusion — Cormack, G. V., Clarke, C. L. A., & Büttcher, S. (2009). "Reciprocal Rank Fusion outperforms Condorcet and individual Rank Learning Methods." SIGIR '09.
Core fusion algorithm for merging results across retrieval layers.
BM25 — Robertson, S. E., Walker, S., Jones, S., Hancock-Beaulieu, M., & Gatford, M. (1994). "Okapi at TREC-3." TREC-3.
Keyword search foundation via SQLite FTS5.
Embeddings & Neural IR
Sentence-BERT — Reimers, N., & Gurevych, I. (2019). "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks." EMNLP 2019. paper
Dense embedding architecture using all-MiniLM-L6-v2.
MiniLM — Wang, W., Wei, F., Dong, L., Bao, H., Yang, N., & Zhou, M. (2020). "MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers." NeurIPS 2020. paper
Efficient transformer distillation for production embedding models.
Reranking & Neural Models
Cross-Encoder Reranking — Nogueira, R., & Cho, K. (2019). "Passage Re-ranking with BERT." arXiv:1901.04085. paper
Cross-attention reranking with TinyBERT on MS MARCO.
TinyBERT — Jiao, X., et al. (2020). "TinyBERT: Distilling BERT for Natural Language Understanding." Findings of EMNLP 2020. paper
Compressed BERT for fast reranking inference.
Vector Search & Systems
FAISS — Johnson, J., Douze, M., & Jégou, H. (2019). "Billion-scale similarity search with GPUs." IEEE Transactions on Big Data. paper
High-performance vector similarity search engine.
GLiNER — Zaratiana, U., Nzeyimana, A., & Holat, P. (2023). "GLiNER: Generalist Model for Named Entity Recognition using Bidirectional Transformer." arXiv:2311.08526. paper
Generalist NER for knowledge graph entity extraction (optional dependency).
Noch keine Tool-Liste. Remote-Server erhalten eine nach dem ersten Health-Check.
FAQ
Fragen
Wie installiere ich VelociRAG?
Kopieren Sie auf dieser Seite die stdio-Konfiguration für Cursor, Claude Desktop, Claude Code oder VS Code. Sie nutzt die Installation des Herausgebers, keinen PlugMCP-Proxy.
Benötigt VelociRAG einen API-Schlüssel?
In diesem Eintrag ist keine erforderliche Umgebungsvariable aufgeführt.
Welchen Transport verwendet VelociRAG?
VelociRAG läuft lokal über stdio.
Wann wurde VelociRAG zuletzt geprüft?
Zuletzt geprüft: 2026-10-27. Lightning-fast RAG for AI agents. ONNX-powered, 4-layer fusion, MCP server. No PyTorch.