RinAI agentic chat stack. Includes graphRAG memory, parallel tool calling, DeepSeek R1 powered web search, advanced LLM gateway and active context summarization.
# RinAI: Your Advanced Agentic Companion

RinAI is a sophisticated agentic companion, leveraging graph-based Retrieval Augmented Generation (RAG), real-time tool usage, intelligent context management, and a dynamic Large Language Model (LLM) gateway. This combination empowers RinAI to engage in roleplaying, hold flirty conversations, and provide insightful responses based on a rich understanding of context and access to up-to-date information.
## Key Features

* **Rin Chat Agent:**
* Engaging roleplay and flirty conversations powered by a fine-tuned uncensored LLM.
* Parallel execution of tools, RAG, and LLM calls for efficient and comprehensive responses.
* Dynamic Smart LLM Gateway intelligently selects the most appropriate model for each query.
* **Graph RAG:**
* Utilizes a Neo4j database containing approximately 18,000 messages, analyzed and indexed for optimal retrieval.
* Retrieval pipeline employs semantic embeddings, sentiment analysis, and subject classification.
* Hybrid search combines vector lookups and rating-based filtering for precise information retrieval.
* **Tool Orchestrator:**
* Integrates with the DeepSeek API (powered by DeepSeek R1) for advanced web search, enabling access to current events, specialized knowledge, and critical reasoning.
* Provides real-time and historical cryptocurrency price checks and analytics via the CoinGecko API.
* Easily extensible architecture allows for seamless integration of new tools by adding API clients.
* **Smart Context Management & Summarization:**
* Automated summarization of conversations once a token threshold is reached.
* Maintains the latest 25% of messages intact while summarizing the older 75% in the background, ensuring contHAL 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。
An LLM agent fine-tuned on DeepSeek for spaced repetition, dynamically integrating knowledge points based on the Ebbinghaus forgetting curve.
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。
A Meta-Agent-Driven Self-Evolving Multi-Agent System for UAV Detection and Tracking
One command to run Hermes AI Agent with a browser UI. Zero prerequisites. 一行命令,AI 就位。
网页应用Agent,接入DeepSeek、Mimo等模型