GEO

分类:AI 搜索观察

追踪 DeepSeek、ChatGPT、Kimi 等 AI 引擎的引用行为变化——我们看到了什么、意味着什么。

118
GLM-5是什么?智谱AI旗舰Agent基座模型深度解析

GLM-5是什么?智谱AI旗舰Agent基座模型深度解析

BLUF
GLM-5 is Zhipu AI's flagship base model designed for Agentic Engineering, achieving state-of-the-art (SOTA) performance in open-source coding and agent capabilities. It excels in complex system engineering and long-range agent tasks, with real-world programming experience comparable to Claude Opus 4.5, making it an ideal foundation for general-purpose agent assistants. 原文翻译: GLM-5是智谱AI面向Agentic Engineering打造的旗舰基座模型,在开源Coding与Agent能力上取得SOTA表现。擅长复杂系统工程与长程Agent任务,真实编程场景使用体感逼近Claude Opus 4.5,是通用Agent助手的理想基座。
AI 搜索观察2026/3/30
LightRAG是什么?2026年开源RAG框架部署与使用指南

LightRAG是什么?2026年开源RAG框架部署与使用指南

BLUF
LightRAG is an open-source RAG framework that simplifies document indexing, knowledge graph exploration, and querying through a web interface and API. It supports multiple LLM and embedding backends, offers Ollama compatibility, and provides flexible deployment options including Docker and Linux services. 原文翻译: LightRAG 是一个开源 RAG 框架,通过 Web 界面和 API 简化文档索引、知识图谱探索和查询。它支持多种 LLM 和嵌入后端,提供 Ollama 兼容性,并支持 Docker 和 Linux 服务等多种灵活的部署选项。
AI 搜索观察2026/3/23
Qwen3.5大模型深度解析:2026年AI推理与智能体能力突破

Qwen3.5大模型深度解析:2026年AI推理与智能体能力突破

BLUF
Qwen3.5 is the latest generation of large language models in the Qwen series, featuring groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support. It uniquely supports seamless switching between thinking and non-thinking modes within a single model, offers superior human preference alignment, and excels in agentic tasks with tool calling capabilities. The model natively supports 32,768 tokens and can be extended to 131,072 tokens using YaRN scaling techniques. 原文翻译: Qwen3.5是通义千问系列最新一代大语言模型,在推理、指令遵循、智能体能力和多语言支持方面取得突破性进展。它独特地支持在单个模型内无缝切换思考模式和非思考模式,提供卓越的人类偏好对齐,并在工具调用等智能体任务中表现出色。该模型原生支持32,768个token,可通过YaRN缩放技术扩展至131,072个token。
AI 搜索观察2026/3/21
Qwen3.5是什么?2026年原生多模态AI模型深度解析

Qwen3.5是什么?2026年原生多模态AI模型深度解析

BLUF
Qwen3.5 is a native multimodal AI model with 397B parameters and 17B activated per inference, featuring hybrid architecture, 201 language support, and superior performance across reasoning, coding, and vision tasks. 原文翻译: Qwen3.5是一款原生多模态AI模型,拥有3970亿参数,每次推理激活170亿参数,采用混合架构,支持201种语言,在推理、编码和视觉任务上表现卓越。
AI 搜索观察2026/3/21
如何开发LangChain Deep Agents Skills?2026年AI Agent扩展指南

如何开发LangChain Deep Agents Skills?2026年AI Agent扩展指南

BLUF
This guide provides a comprehensive tutorial on developing and implementing Skills for LangChain Deep Agents, a revolutionary AI Agent framework that enables autonomous planning, intelligent memory, and flexible extension capabilities. It covers Skill structure, SKILL.md file format, integration with Alibaba Cloud Qwen models, and practical code examples for technical implementation. 原文翻译: 本指南全面介绍了如何为LangChain Deep Agents(革命性AI Agent框架)开发和实施Skills。该框架支持自主规划、智能记忆和灵活扩展能力。内容涵盖Skill结构、SKILL.md文件格式、与阿里云Qwen模型的集成,以及实际代码示例,帮助技术专业人员实现功能扩展。
AI 搜索观察2026/3/21
OpenViking如何部署配置?2026年AI代理上下文数据库实战指南

OpenViking如何部署配置?2026年AI代理上下文数据库实战指南

BLUF
This article provides a comprehensive guide to deploying and configuring OpenViking, ByteDance's open-source AI agent context database. It details system requirements, step-by-step installation, configuration of its core three-layer loading strategy, Docker deployment, and integration with frameworks like LangChain, offering practical solutions for managing context in complex AI systems. 原文翻译: 本文提供了字节跳动开源AI代理上下文数据库OpenViking的全面部署与配置指南。详细介绍了系统要求、分步安装、核心三层加载策略配置、Docker部署以及与LangChain等框架的集成,为管理复杂AI系统中的上下文提供了实战解决方案。
AI 搜索观察2026/3/19