GitHub – nilsherzig/LLocalSearch: LLocalSearch is a completely locally running search aggregator using LLM Agents. The user can ask a question and the system will use a chain of LLMs to find the answer. The user can see the progress of the agents and the final answer. No OpenAI or Google API keys are needed.

4个月前更新 97 00

LLocalSearch is a completely locally running search aggregator using LLM Agents. The user can ask a question and the system will use a chain of LLMs to find the answer. T...

所在地:
中国
语言:
zh
收录时间:
2025-04-05
其他站点:
GitHub – nilsherzig/LLocalSearch: LLocalSearch is a completely locally running search aggregator using LLM Agents. The user can ask a question and the system will use a chain of LLMs to find the answer. The user can see the progress of the agents and the final answer. No OpenAI or Google API keys are needed.GitHub – nilsherzig/LLocalSearch: LLocalSearch is a completely locally running search aggregator using LLM Agents. The user can ask a question and the system will use a chain of LLMs to find the answer. The user can see the progress of the agents and the final answer. No OpenAI or Google API keys are needed.

LLocalSearch:本地化智能搜索聚合解决方案

▎核心功能亮点
全离线架构
基于Ollama本地LLM框架构建,彻底消除云服务数据泄露风险。通过ollama run mistral等指令即可部署私有化模型。

代理链式工作流
采用三级代理协同机制(问题解析→搜索规划→结果验证),通过agent_count参数动态调节并发规模,支持最高12线程并行处理。

透明化进程追踪
Gradio可视化界面实时展示代理思考路径,开发者可通过verbose参数获取完整调试日志。

▎技术优势解析
1. 零API依赖设计
对比传统LangChain方案,彻底摆脱对OpenAI/Gemini等商业API的依赖,符合GDPR企业级部署规范。

2. 模块化扩展架构
支持通过custom_agents目录添加自定义代理,兼容HuggingFace模型库(需5GB+显存配置)。

3. 混合检索增强
集成BM25算法与语义搜索,通过rerank_threshold=0.8参数平衡精准度与召回率。

▎典型应用场景
✅ 企业内网知识库检索
✅ 医疗/金融领域敏感数据查询
✅ 物联网设备离线问答系统
✅ AI代理机制教学实验平台

bash
快速启动指令
git clone https://github.com/nilsherzig/LLocalSearch
python main.py model mistral agent_count 4

开源地址:https://github.com/nilsherzig/LLocalSearch
(项目采用MIT协议,支持Docker/Kubernetes集群化部署)

相关导航

暂无评论

您必须登录才能参与评论!
立即登录
none
暂无评论...