React Refs made easy.

演示地址：[el-rrf-vue-element-base-components]

Postgres and pgvector adapter for @jeffs-brain/memory: tsvector BM25, halfvec cosine, and RRF hybrid retrieval.

MemberJunction: Reusable server-side search engine - vector + full-text + entity search with RRF fusion

TeamBrain hybrid retrieval: SQLite FTS5 + sqlite-vec + RRF behind a RetrievalBackend interface

Local-first recall engine: BM25 + multilingual-e5-small + RRF

MCP server giving AI agents (Claude Code, Claude Desktop, Cursor, ChatGPT, Codex, OpenClaw) persistent long-term memory backed by your local Obsidian markdown vault. Hybrid retrieval (BM25 + ML embeddings + BGE reranker, RRF-fused), HNSW + int8 quantizati

OpenCodeHub — BM25 + RRF hybrid search

Retrieval: chunker, BM25 + dense hybrid with RRF, schema-bound citation validator.

Persistent memory for AI agents. Multi-strategy retrieval (semantic + BM25 + RRF), 3-zone cosine triage, type-aware lifecycle, entity canonicalization. Postgres + pgvector.

Hybrid keyword + embedding semantic matcher with RRF fusion

Adapter-agnostic core for Agentic Task System — retrieval (RRF parallel fan-out), corpus cache, usage logging, conformance kit, adapter interface

Model Context Protocol server for Agentic Task System — exposes the task app you already use to any MCP client, backed by hybrid + RRF retrieval. Storage-agnostic over the ATS adapter contract.

Local-first recall engine: BM25 + multilingual-e5-small + RRF

RyoForge Runtime Framework (RRF) - A CLI-driven, metadata-driven, method-oriented backend framework for Node.js & Express. Ships the `rrf` command and the runtime library.

Weighted, adaptive, calibration-free reciprocal-rank fusion (RRF) of heterogeneous retrieval channels.

Semantic code-search MCP installer and indexer for AI coding agents: hybrid (vector + lexical RRF) ranking, gitignore-aware indexing, and optional cross-encoder rerank.

Cognitive core algorithms (RRF fusion, SMG gating, surprisal detection) for Memo

High-performance local RAG for Markdown: contextual FTS5, HNSW vector search, weighted RRF, and on-device ONNX or llama.cpp embeddings. Bun-only.

Self-improving memory engine for Eidentic — four-tier recall (lexical + semantic RRF), self-editing blocks, temporal graph, and sleep-time consolidation.