8.2 LlamaIndex核心组件 文档分割向量与查询引擎

8.2 LlamaIndex 核心组件：文档、分割、向量、查询引擎

本节学习目标

• 掌握 LlamaIndex 的文档加载、文本分割、向量索引、查询引擎在 RAG 管线中的角色。
• 能说出每步的常用类与配置（如 chunk 大小、overlap、embedding 模型）。
• 为 8.3 从零构建知识库 Agent 打基础。

一、组件链

#mermaid-svg-utk9KArI0o8lDOzu{font-family:\”trebuchet ms\”,verdana,arial,sans-serif;font-size:16px;fill:#333;}@keyframes edge-animation-frame{from{stroke-dashoffset:0;}}@keyframes dash{to{stroke-dashoffset:0;}}#mermaid-svg-utk9KArI0o8lDOzu .edge-animation-slow{stroke-dasharray:9,5!important;stroke-dashoffset:900;animation:dash 50s linear infinite;stroke-linecap:round;}#mermaid-svg-utk9KArI0o8lDOzu .edge-animation-fast{stroke-dasharray:9,5!important;stroke-dashoffset:900;animation:dash 20s linear infinite;stroke-linecap:round;}#mermaid-svg-utk9KArI0o8lDOzu .error-icon{fill:#552222;}#mermaid-svg-utk9KArI0o8lDOzu .error-text{fill:#552222;stroke:#552222;}#mermaid-svg-utk9KArI0o8lDOzu .edge-thickness-normal{stroke-width:1px;}#mermaid-svg-utk9KArI0o8lDOzu .edge-thickness-thick{stroke-width:3.5px;}#mermaid-svg-utk9KArI0o8lDOzu .edge-pattern-solid{stroke-dasharray:0;}#mermaid-svg-utk9KArI0o8lDOzu .edge-thickness-invisible{stroke-width:0;fill:none;}#mermaid-svg-utk9KArI0o8lDOzu .edge-pattern-dashed{stroke-dasharray:3;}#mermaid-svg-utk9KArI0o8lDOzu .edge-pattern-dotted{stroke-dasharray:2;}#mermaid-svg-utk9KArI0o8lDOzu .marker{fill:#333333;stroke:#333333;}#mermaid-svg-utk9KArI0o8lDOzu .marker.cross{stroke:#333333;}#mermaid-svg-utk9KArI0o8lDOzu svg{font-family:\”trebuchet ms\”,verdana,arial,sans-serif;font-size:16px;}#mermaid-svg-utk9KArI0o8lDOzu p{margin:0;}#mermaid-svg-utk9KArI0o8lDOzu .label{font-family:\”trebuchet ms\”,verdana,arial,sans-serif;color:#333;}#mermaid-svg-utk9KArI0o8lDOzu .cluster-label text{fill:#333;}#mermaid-svg-utk9KArI0o8lDOzu .cluster-label span{color:#333;}#mermaid-svg-utk9KArI0o8lDOzu .cluster-label span p{background-color:transparent;}#mermaid-svg-utk9KArI0o8lDOzu .label text,#mermaid-svg-utk9KArI0o8lDOzu span{fill:#333;color:#333;}#mermaid-svg-utk9KArI0o8lDOzu .node rect,#mermaid-svg-utk9KArI0o8lDOzu .node circle,#mermaid-svg-utk9KArI0o8lDOzu .node ellipse,#mermaid-svg-utk9KArI0o8lDOzu .node polygon,#mermaid-svg-utk9KArI0o8lDOzu .node path{fill:#ECECFF;stroke:#9370DB;stroke-width:1px;}#mermaid-svg-utk9KArI0o8lDOzu .rough-node .label text,#mermaid-svg-utk9KArI0o8lDOzu .node .label text,#mermaid-svg-utk9KArI0o8lDOzu .image-shape .label,#mermaid-svg-utk9KArI0o8lDOzu .icon-shape .label{text-anchor:middle;}#mermaid-svg-utk9KArI0o8lDOzu .node .katex path{fill:#000;stroke:#000;stroke-width:1px;}#mermaid-svg-utk9KArI0o8lDOzu .rough-node .label,#mermaid-svg-utk9KArI0o8lDOzu .node .label,#mermaid-svg-utk9KArI0o8lDOzu .image-shape .label,#mermaid-svg-utk9KArI0o8lDOzu .icon-shape .label{text-align:center;}#mermaid-svg-utk9KArI0o8lDOzu .node.clickable{cursor:pointer;}#mermaid-svg-utk9KArI0o8lDOzu .root .anchor path{fill:#333333!important;stroke-width:0;stroke:#333333;}#mermaid-svg-utk9KArI0o8lDOzu .arrowheadPath{fill:#333333;}#mermaid-svg-utk9KArI0o8lDOzu .edgePath .path{stroke:#333333;stroke-width:2.0px;}#mermaid-svg-utk9KArI0o8lDOzu .flowchart-link{stroke:#333333;fill:none;}#mermaid-svg-utk9KArI0o8lDOzu .edgeLabel{background-color:rgba(232,232,232, 0.8);text-align:center;}#mermaid-svg-utk9KArI0o8lDOzu .edgeLabel p{background-color:rgba(232,232,232, 0.8);}#mermaid-svg-utk9KArI0o8lDOzu .edgeLabel rect{opacity:0.5;background-color:rgba(232,232,232, 0.8);fill:rgba(232,232,232, 0.8);}#mermaid-svg-utk9KArI0o8lDOzu .labelBkg{background-color:rgba(232, 232, 232, 0.5);}#mermaid-svg-utk9KArI0o8lDOzu .cluster rect{f