mmcp

Name: mmcp
Author: RagavRida

RagavRida February 28, 2026

32 copies 0 downloads

Multi-Model Collaboration Pipeline — orchestrate AI models as a DAG. RL routing, multi-verifier voting, agent mesh, self-improving. Works with OpenAI, Anthropic, Gemini, DeepSeek. npm install mmcp-core | pip install mmcp-core

Agent Definition

<div align="center">

# 🔀 MMCP — Multi-Model Collaboration Pipeline

**Orchestrate AI models as a coordinated DAG. RL routing · Multi-verifier voting · Agent mesh · Self-improving.**

[![npm](https://img.shields.io/npm/v/mmcp-core?style=flat-square&logo=npm&color=red)](https://npmjs.com/package/mmcp-core)
[![PyPI](https://img.shields.io/pypi/v/mmcp-core?style=flat-square&logo=pypi&label=PyPI&color=blue)](https://pypi.org/project/mmcp-core/)
[![Downloads](https://img.shields.io/pypi/dm/mmcp-core?style=flat-square&label=downloads)](https://pypi.org/project/mmcp-core/)
[![License](https://img.shields.io/badge/license-MIT-blue?style=flat-square)](LICENSE)
[![Railway](https://img.shields.io/badge/API-live-brightgreen?style=flat-square&logo=railway)](https://mmcp.up.railway.app/health)
[![GitHub stars](https://img.shields.io/github/stars/RagavRida/mmcp?style=flat-square&logo=github)](https://github.com/RagavRida/mmcp/stargazers)

</div>

---

<div align="center">

**MCP standardizes tool use for a single model.**
**MMCP standardizes context flow *between* models.**

</div>

---

## ⚡ 30-Second Quick Start

```bash
pip install mmcp-core
mmcp login
mmcp run
```

That's it. Type a task, MMCP picks the best model + pattern automatically.

## 🧠 Domain-Aware RL Routing — The Right Model for Every Task

MMCP doesn't just pick a model — it **learns per domain** which model performs best, then routes automatically. When models get updated, benchmark results feed back into the router.

| Your Task | Domain Detected | Model Selected | Domain Score |
|-----------|----------------|---------------|-------------|
| "Write a Python API with auth" | `code_generation` | **GPT-4o** | 0.96 |
| "Debug this React component" | `code_review` | **Claude Sonnet** | 0.91 |
| "Prove this calculus theorem" | `math_reasoning` | **DeepSeek R1** | 0.92 |
| "Write a blog post about AI" | `creative_writing` | **Claude Sonnet** | 0.90 |
| "Find SQL injection in this code" | `security` | **Claude Opus** |

Comments

More Agents

View all

hybrid-model-workflow

HAL 分层混合模型工作流 — 强模型(Claude)负责理解/拆解/验收，低成本模型(DeepSeek)负责检索/提取/清洗。Hermes Agent skill。

ph4ble

Dynamic-Review-Agent

An LLM agent fine-tuned on DeepSeek for spaced repetition, dynamically integrating knowledge points based on the Ebbinghaus forgetting curve.

1838177

StellarOS-Watch

基于 STM32F103 构建的端到端 AI 智能手表生态。自研“零重定位”原生机器码动态加载引擎与页面栈式 UI 框架；集成生产级 OTA 回滚保护机制与高带宽（921600 baud）串口协议栈。通过 Node.js 中继实现 DeepSeek AI 语义控制及 ASRPRO 语音全双工交互，是一个集成了分布式计算、现代存储管理与 AI Agent 的嵌入式全栈工程。

chenshuang888