大模型
x.com ·
@llama_index
LlamaIndex 详解 just-in-time OCR：LiteParse 首遍 + LlamaParse 精读
LlamaIndex 官方介绍 just-in-time OCR 两段式流程：LiteParse（免费、开源、Rust 实现、支持 50+ 格式）做快速布局感知首遍，产出空间文本、边界框、标题、表格及逐页复杂度标记，官方称完整数据室可在 32 秒内完成；LlamaParse 随后按页号仅精读需要的页面，返回单元格级表格、边界框与置信度分数，其余页面在后台补齐。帖子称 pypdf 与 pdftotext 难以胜任首遍，并链接博客给出具体数字。
AI推荐理由 LlamaIndex 官方账号给出 just-in-time OCR 两段式方案的具体分工：LiteParse（免费、OSS、Rust、支持 50+ 格式）做布局感知首遍，输出空间文本、边界框、标题、表格和逐页复杂度标记，称完整数据室 32 秒完成；LlamaParse 按页号精读需要的页面，返回单元格级表格、边界框和置信度。有明确工具与流程，但性能数字来自自家博客，且为产品推广，需保留其未独立验证的性质。
68
浏览 1,656
点赞 10
评论 3
转发 2
正文
just-in-time OCR is all the rage.
most pipelines parse every page before anyone asks a question. for an agent working through an ad-hoc data room, that's slow, expensive, and most of those pages never get read.
the better pattern is just-in-time OCR in two passes:
✅️ LiteParse (free, OSS, Rust, 50+ formats) does a fast layout-aware first pass: spatial text, bounding boxes, headings, tables, and a per-page complexity flag. a full data room in 32 seconds.
✅️ LlamaParse zooms in on only the pages that need it, by page number, and returns cell-level tables, bounding boxes, and confidence scores. the rest fills in the background.
pypdf and pdftotext can't do the first pass well. parsing everything up front can't do it cheaply. two passes gets you both.
full breakdown with numbers: [llamaindex.ai/blog/just-in-t…](https://www.llamaindex.ai/blog/just-in-time-agentic-ocr)
