MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
Convert ONNX models to Apple Core AI (.aimodel) for on-device inference on Apple silicon. 143 op lowerings, precision checks, Python API, CLI, and MCP.
BSD-3-Clause ★ 5 -
Local-first MCP server for persistent semantic memory, backed by an HNSW vector index and an on-device ONNX embedder
MIT -
qa-memorize-mcpupdated this year
RAG-based memory MCP server with local ONNX embeddings
npm MIT -
piper-mcp-servergithub.com/nizovtsevnv/piper-mcp-serverupdated this year
Text-to-speech MCP server using Piper ONNX models
cargo MIT -
本地知识库(ONNX 中文向量 + MCP 服务 + DSH 插件):不依赖外部服务,聊天时自动检索注入
MIT -
轻量向量知识库工具 — 本地 ONNX 嵌入 + sqlite-vec 混合检索 + MCP 工具接口 + RAG 对话
AGPL-3.0 ★ 2 -
Offline voice control for OpenCode: push-to-talk ASR (SenseVoice + sherpa-onnx) with MCP tools
MIT -
memory-safegithub.com/eljoesb/memory-safeupdated this year
MCP memory server for Claude Code — knowledge graph in SQLite, local ONNX embeddings, token-budgeted retrieval
npm MIT -
ONNX Runtime stem separation (Demucs HTDemucs). Separate vocals, drums, bass, other from any audio file. Model auto-downloads from HuggingFace. Companion to mixx-dj-mcp.
MIT ★ 1 -
Local-first embedding server (ONNX on-device or API) — the reference /embed server for CPersona. MIT.
pypi MIT ★ 1 -
MCP server for multilingual place/location extraction (cities, countries, regions, landmarks) in 13 languages, backed by mDeBERTa/ONNX models.
MIT -
Agent-Speakergithub.com/NetDevAutomate/Agent-Speakerupdated this year
A small mcp/tool for agents (Kiro CLI, Claude Code and others...) to speak responses using a wrapper for kokoro-onnx (https://github.com/thewh1teagle/kokoro-onnx) — an 82M parameter ONNX TTS model
★ 6 -
Self-hosted real-time X stream intelligence: DAG pipelines of local ONNX models + Ollama LLMs, live map, entity graph, alerts, MCP server. Go + Vue.
MIT -
Native Rust AI orchestration CLI — zero Node.js dependency. 347 MCP tools. SONA neural net, Thompson bandit router, RuVector HNSW, real ONNX embeddings.
-
Long-term memory for AI assistants, local. Rust + Qdrant + ONNX, served over MCP. Hybrid search with reranking.
npm Apache-2.0 ★ 10 -
Rust-native hybrid retrieval over LLM thinking traces - dense ANN + BM25 + rerank, all local ONNX. Daemon, HTTP + MCP, zero API keys.
Apache-2.0 ★ 1 -
rag-localgithub.com/MauricioPerera/rag-localupdated this year
Local-first RAG memory for LLMs: Node server with ONNX embeddings, OKF-structured chunks, portable vector bundles, and REST + Web UI + CLI + MCP interfaces
MIT -
LEANN MCP Server — semantic code search powered by ONNX Runtime. Provides MCP tools for RAG-based code retrieval with GPU-accelerated embedding (DirectML on Windows, CoreML on macOS).
nuget MIT -
Semantic code search for AI agents and CLI users. CLI + MCP server using local ONNX embeddings — no API keys, no external services.
nuget ★ 1 -
A high-performance local knowledge system for Claude Code via MCP. PostgreSQL + pgvector + ONNX embeddings. Gives your AI assistant persistent memory across sessions.
MIT ★ 1 -
MCP server in Rust for AI agent persistent memory: branch-aware session handoffs, local ONNX embeddings, SQLite-backed semantic search.
cargo Apache-2.0 ★ 19 -
MCP server and MLOps CLI to export, quantize, audit, benchmark and auto-tune PyTorch and ONNX computer-vision models for edge deployment.
MIT -
Rust MCP server for generative media with avatars: InsightFace/ArcFace identity core, ONNX face swap, budget-guarded pipeline. Work in progress.
NOASSERTION -
A lightweight, containerized multi-agent platform featuring an intelligent Router Agent, pre-built & dynamic sub-agents, FastMCP tool integrations, ONNX-powered vector RAG, and PostgreSQL chat persistence.
★ 1 -
? Tiny Decision - A Local Harness & Deployment Platform for System One Models. In-process GGUF/ONNX runtime, calibrated confidence, visual DAG pipelines, and MCP export.
MIT ★ 1 -
DeepFetchgithub.com/vinay9986/DeepFetchupdated this year
Semantic web search MCP server with Kagi, Scrapfly, PDF extraction, and local ONNX reranking for Claude Desktop, Gemini CLI, and Codex.
Apache-2.0 -
JS-PotatoRAGgithub.com/MauricioPerera/JS-PotatoRAGupdated this year
Consola RAG local air-gapped + memoria para agentes (MCP). Base vectorial cuantizada en 3 bits (Rust/WASM, ~21x), embeddings ONNX y LLM locales, PWA instalable.
MIT -
descrygithub.com/jKiler/descryupdated this year
Local hybrid (vector + BM25) code search for AI coding agents — AST chunking, MiniLM embeddings on ONNX, Go call graph, MCP server. No external services.
go Apache-2.0 ★ 1 -
Production embedding & decision inference service in Rust — Candle/ONNX dual engine, CUDA/Metal acceleration, REST/gRPC/MCP/CLI generated via sdforge, rerank, Matryoshka dimensions, OpenAI-compatible API, and semantic caching
Apache-2.0 ★ 1 -
Local, bilingual (ES/EN) semantic search over a technical book library for coding assistants: SQLite FTS5 + ONNX embeddings, a fail-open prompt hook and an MCP server. Includes a chunking and embedding-model study (docs/paper.md).
Apache-2.0 -
Static security scanner for machine-learning model files. Checks PyTorch, pickle, joblib, GGUF, SafeTensors, ONNX and Keras checkpoints for code execution before you load them. CLI, GitHub Action, load guard, MCP server.
pypi MIT ★ 2 -
MCP server with 40 tools for web design analysis — layout extraction, motion detection, quality scoring, accessibility audit, Core Web Vitals, design comparison, and semantic search via Playwright, pgvector, ONNX Runtime, and Ollama.
npm AGPL-3.0 ★ 2 -
Persistent, local-first memory for AI agents. Selective by design: filters what's worth keeping, links what's related, and says nothing when it doesn't know. SQLite + ONNX, no cloud, no API keys.
npm Apache-2.0 ★ 4 -
Free, open-source desktop attention prediction (Windows & macOS). Predict where people look in the first ~3s of free viewing. 100% offline, private ONNX inference (SUM, UNISAL, TranSalNet). Transparent research-backed metrics (AOI share, clarity). GUI, CLI, local HTTP API, and MCP agent server.
MIT -
Long-term memory plugin for DeepSeek Harness (MCP): L1 fact cards + L2 evidence timeline + L3 dialogue log + L4 episode summaries, 21 memory_* tools, per-turn auto weave/inject and scoped ingest, hybrid semantic+BM25+recency RRF retrieval, AES-256-GCM field encryption, local ONNX embeddings, zero API cost.
Apache-2.0 -
A desktop phone for your coding agent - it calls you over MCP and you have a natural, spoken back-and-forth. Voice runs fully on-device by default (sherpa-onnx: Parakeet STT + Kokoro TTS), with optional ElevenLabs. Built with Tauri v2 (Rust) + React.
MIT ★ 1 -
A local semantic-recall MCP server for citation-integrity academic writing — fastembed ONNX embeddings on LanceDB, cross-lingual, fully offline. Recall-only by contract: fragments carry file and page; the source PDF supplies every quote. Companion to alexandria. 服務引文完整性寫作的本地語意召回 MCP 伺服器:跨語言、完全離線;只召回不引用,片段附檔案與頁碼,每一句引文都回原始 PDF 取。
MIT ★ 1 -
Законы и кодексы, mcp, cli, qwen3-emb-onnx
BSD-2-Clause -
tts-apigithub.com/subconscious-labs/tts-apiupdated this year
Kokoro ONNX based TTS API/MCP
MIT ★ 2 -
Local AI voice for coding assistants — TTS & STT via MCP. Kokoro ONNX + faster-whisper, fully offline.
npm Apache-2.0 ★ 11 -
mcp-server-qdrantgithub.com/vaporif/mcp-server-qdrantupdated this year
Rust MCP server for Qdrant with local embeddings (Candle/ONNX)
cargo MIT -
面向软件实施与技术支持人员的 Java Agent 诊断工作台:只读 MCP、版本感知 RAG、文档与 Excel 处理、可选 ONNX 重排序及可追溯诊断。
Apache-2.0 -
memexgithub.com/ayushagrawal288/memexupdated this year
Production-grade persistent memory service for AI agents — FastAPI + pgvector + local ONNX embeddings
oci MIT ★ 1 -
Episodic memory for AI agents: structured episode storage, semantic recall on ChromaDB + ONNX MiniLM, exposed over MCP.
MIT ★ 1 -
This is a local LLM provider, include rag-mcp and agent features, can load HF/gguf/onnx model formats..
Apache-2.0 -
annalgithub.com/heyhayes/annalupdated this year
Semantic memory server for AI agent teams. MCP server, ChromaDB or Qdrant, local ONNX embeddings.
MIT -
Rust code graph indexer with 28+ language support, MCP tools, ONNX vector embeddings, and SQLite-backed semantic analysis
NOASSERTION ★ 15 -
MMPL (Memory Palace) Rust Edition — 单二进制 DuckDB 记忆 MCP Server (9 工具 + ONNX 语义向量 + 冲突双通道门控 + 场景构建 + 四层钻取) — Linux + Windows 双平台
npm MIT -
mcp-rag-minigithub.com/Pinger1456/mcp-rag-miniupdated this year
Minimal RAG service exposing one ChromaDB index via both REST (FastAPI) and MCP (stdio). Python 3.12 · fastembed ONNX.
MIT -
MMPL (Memory Palace) — 单二进制 DuckDB 记忆 MCP Server(10 工具:记忆 L0-L3 + 技能库 + ONNX 语义向量 + 遗忘曲线 + 四层钻取)— Linux + Windows 双平台
npm MIT -
树莓派上的果蔬计价秤:识别、称重、计价、语音播报、小票打印,同一套能力以命令行、LLM function calling 和 MCP 三种方式暴露。MobileNetV3 + ONNX,实拍留出集 88.5%
MIT -
VelociRAGgithub.com/HaseebKhalid1507/VelociRAGupdated this year
Lightning-fast RAG for AI agents. ONNX-powered, 4-layer fusion, MCP server. No PyTorch.
pypi MIT ★ 12 -
kokoro-tts-kotlingithub.com/alexsobolev/kokoro-tts-kotlinupdated this year
Production ready text to speech service built with Kotlin, Kokoro (82M parameters), ONNX Runtime, and Clean Architecture. REST API + MCP server for AI assistants. Multi speaker dialogue, POS aware phonemization, AWS deployment (EC2/Lambda).
Apache-2.0 ★ 10 -
@adamrdrew/agent-memory-mcpgithub.com/adamrdrew/agent-memory-mcpupdated this year
MCP server for persistent agent memory — hybrid BM25 + vector search backed by LanceDB with local ONNX embeddings. No API keys, fully local.
npm GPL-3.0-or-later -
Yggdrasilgithub.com/IT-Square-Plus/Yggdrasilupdated this year
MCP memory server for Claude Code with semantic search. Chroma Cloud storage, ONNX embeddings, path-based routing. One container serves multiple projects with isolated memory spaces. FastMCP + FastAPI.
MIT ★ 1 -
argusgithub.com/maisymylod/argusupdated this year
Natural-language earth-observation: an LLM agent calls geospatial tools (also exposed over MCP), runs an NDVI/ONNX imagery pipeline on public Sentinel-2 data, and renders to a WebGL map.
MIT -
Researchmind_MCP_Servergithub.com/suraj5424/Researchmind_MCP_Serverupdated this year
ResearchMind MCP Server gives AI assistants a local research brain with RAG, semantic search, arXiv integration, and persistent memory. Built with Python, ONNX Runtime, ChromaDB, and SQLite.
MIT -
The Pure Rust Model Context Protocol (MCP) server for AI coding agents. Grounding swarms in deterministic AST graphs, BM25, and local ONNX vectors.
MIT -
Privacy-first Multimodal Data Refinery, Repo Context Firewall (Prompt Injection & Trojan Source Defense), Offline Vietnamese OCR, Multilingual E5 ONNX + SQLite FTS5 RAG & 9-Tool MCP Server.
MIT ★ 7 -
Local AI memory for coding agents: MCP server with hybrid BM25 + vector search, session capture, stale-memory prune, and git-committable BATTERY.md. SQLite + ONNX embeddings—no cloud DB or embed APIs.
MIT