Local Deep Research:终极本地优先 AI 深度研究工具
掌握 Local Deep Research (LDR) —— 本地优先的 AI 研究助手。了解如何结合 Ollama 和 SearXNG 进行深度迭代研究,同时保持 100% 的隐私。
- Docker
- Python
- MIT
- 更新于 2026-05-15
🔗 相关资源推荐
AI Agent Tool Chain 2026: The 6-Component Stack for Building Production-Grade Autonomous Agents
Complete production AI agent stack: LangGraph for stateful orchestration + MCP servers for tools + mem0 for memory + OpenClaw for multi-agent coordination + Hermes Agent for self-improvement + e2b for sandboxed code execution. $20-60/mo self-hosted. Real assembly with internal-linked deep dives.
Cross-Border AI Marketing Stack 2026: 7-Tool Setup for Chinese Teams Shipping Globally
A 7-component AI stack purpose-built for cross-border operations — automate multilingual content, scrape global market intel, GDPR-compliant analytics, bypass payment friction, and run the whole thing from a Hong Kong VPS. $35-80/mo total, all-OSS-or-aff-friendly.
The AI Trading Stack 2026: 7-Component Open-Source Quant Workflow for Crypto + Prediction Markets
Self-hosted AI trading stack: ta-lib (signals) + vectorbt (backtest) + freqtrade (execution) + AI Trader (AI strategy layer) + Hyperliquid (perp DEX venue) + Polymarket Agents (prediction markets) + Minara (AI+crypto hub). $30-150/mo infrastructure, real production-grade quant pipeline, not a toy.
The Cheap LLM Stack 2026: How to Run Production AI on $0-15/Month Using Free Tiers and Token Compression
5-component stack to run real AI workloads on $0-15/month: Ollama local + DeepSeek API + Gemini free tier + RTK compression + 9Router orchestration. Real cost math, model picks per task type, assembly order.
The Knowledge Base Stack 2026: Build Your "Second Brain" with AnythingLLM + RAGFlow + mem0 ($10-25/Month)
A 5-component self-hosted knowledge base stack for personal or team use. AnythingLLM (UI + RAG) + RAGFlow (deep doc parsing) + mem0 (agent memory) + AgentMemory MCP (MCP exposure) + vector DB pick. Replaces $50-200/mo SaaS (Notion AI + Mem + Glean) with $10-25/mo self-hosted.
MCP Server 全目录指南 2026:19,700+ 服务器、7 个官方款、以及 60 秒挑对那一个
2026 MCP server 发现完全指南。Anthropic 7 个 reference 服务器、87.3k star 的 awesome list、Smithery vs mcp.so 两大注册中心对比、各类别 top 服务器、以及挑选决策树 —— 不讲 MCP 是什么,只讲市面上有什么可以插进你的 MCP host。
💬 留言讨论