MNMA

On-premises conversational RAG with configurable containers

No commits in 8 months — may not be actively maintained. See maintained alternatives →

open-sourcememory-knowledge
Visit WebsiteView on GitHub
1.0k
Stars
+2
Stars/month
0
Commits (90d)
0
Releases (6m)

Star Growth

+10 (1.0%)
1.0k1.0k1.1kMar 27Oct 2

Overview

Minima是一个开源的本地部署RAG（检索增强生成）系统，通过容器化部署提供灵活的文档问答能力。该工具支持四种不同的部署模式：完全本地化部署（使用Ollama）、自定义LLM集成（OpenAI兼容API）、ChatGPT集成以及Anthropic Claude集成。Minima的核心优势在于数据安全性和部署灵活性，用户可以选择完全离线运行以确保数据隐私，也可以结合云端LLM服务获得更强的推理能力。系统包含完整的RAG流水线，支持文档索引、向量检索、重排序和生成等功能。通过Docker容器化部署，Minima简化了复杂的依赖管理和环境配置问题。该工具特别适合企业级应用场景，在需要处理敏感文档同时又要保证查询质量的情况下，提供了理想的解决方案。其模块化设计允许用户根据具体需求选择合适的部署模式，从完全本地化到混合云部署都能很好支持。

Deep Analysis

Key Differentiator

vs cloud RAG (ChatGPT retrieval/Perplexity): four deployment modes from fully local to cloud-integrated, with MCP protocol for IDE integration — data stays on-premises

⚡ Capabilities

  • • On-premises RAG system for document indexing and querying
  • • Supports PDFs, Excel, Word, text, markdown, CSV
  • • Four operational modes: fully local, custom LLM, ChatGPT, MCP/Claude
  • • Vector search with optional reranking (BAAI models)
  • • MCP protocol support for IDE integration (GitHub Copilot)
  • • Recursive directory indexing

🔗 Integrations

ChatGPTAnthropic ClaudeOllamavLLMTGILiteLLMLocalAIQdrantSentence TransformersGitHub Copilot (MCP)

✓ Best For

  • ✓ Organizations needing sensitive document search without cloud exposure
  • ✓ Teams wanting flexible RAG with local-to-cloud deployment spectrum

✗ Not Ideal For

  • ✗ Public-facing document systems
  • ✗ Organizations without local computing resources

Languages

Python (MCP)JavaScript/Electron (desktop)

Deployment

Docker Compose (4 modes)fully local (Ollama)hybrid (local index + cloud LLM)

⚠ Known Limitations

  • ⚠ Limited to 6 document formats (PDF, Excel, Word, txt, md, CSV)
  • ⚠ Embedding restricted to Sentence Transformers architecture
  • ⚠ Custom LLM mode disables reranking
  • ⚠ Requires Python 3.10+ and uv for MCP functionality

Pros

  • + 数据隐私保护 - 支持完全本地部署，确保敏感文档不离开本地环境
  • + 部署模式灵活 - 提供4种不同部署模式，适应不同的技术栈和安全需求
  • + 容器化部署简单 - 通过Docker和一键脚本大幅简化安装和配置流程

Cons

  • - 资源需求较高 - 完全本地部署需要足够的计算资源运行多个神经网络模型
  • - 配置相对复杂 - 多种部署模式需要不同的环境变量和配置文件设置
  • - 依赖Docker环境 - 需要用户具备容器化部署的基础知识

Use Cases

  • • 企业内部文档智能问答 - 在保证数据安全的前提下构建内部知识库检索系统
  • • 个人本地知识管理 - 对本地文档集合进行智能检索和问答，无需上传到云端
  • • 混合RAG架构集成 - 与现有LLM基础设施集成，实现本地索引+云端推理的混合模式

Getting Started

1. 克隆项目并根据.env.sample创建.env配置文件，设置文档路径和模型参数；2. 运行./run.sh脚本，从4种部署模式中选择适合的方案；3. 等待容器启动完成后，开始上传文档并进行智能问答测试

Alternatives

l
localGPT
Chat with your documents on your local device using GPT models. No data leaves your device and 100% private.
⭐ 22.2k
r
ragflow
Open-source RAG engine combining knowledge retrieval and agent capabilities for LLMs
⭐ 91.6k
V
Verba
Retrieval Augmented Generation (RAG) chatbot powered by Weaviate
⭐ 7.7k
C
Canopy
Retrieval Augmented Generation (RAG) framework and context engine powered by Pinecone
⭐ 1.0k
See all 8 MNMA alternatives →

Works with MNMA

Tools that integrate with MNMA, often used together in the same stack.

O
Ollama
Get up and running with Kimi-K2.5, GLM-5, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
v
vLLM
A high-throughput and memory-efficient inference and serving engine for LLMs
L
LiteLLM
Open-source Python SDK and AI gateway for calling 100+ LLMs through a unified OpenAI-compatible interface
L
LocalAI
LocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
Q
Qdrant
Vector similarity search engine and database written in Rust, with payload filtering and managed cloud service

Compare MNMA

MNMA vs HindsightMNMA vs CogneeMNMA vs ragflowMNMA vs GraphifyMNMA vs MinerUMNMA vs headroom

Maintain MNMA?

Show your live rank in your README, or put MNMA in front of every visitor to AgentoolRank.