Submit repository
Discover trends that matter
Daily explore
Live mentions
Topics
GitHub trending
Repositories
Developers
Repository engagements
Insights
Stats
VectifyAI/PageIndex — GitHub trending stats & insights | Trendshift
Featured
Openhuman
Embed Badge
Visit GitHub
VectifyAI/PageIndex
#
RAG
#
Document processing
📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG
Data last synced with GitHub 1 day ago
Python
31.2k
2.7k
12 contributors
last commit 4 days ago
last user commit 4 days ago
MIT License
website
created about 1 year ago
Social mentions
Recent discussions about this repository across the web
这就是本周最热的 GitHub 仓库! 1. DeepSeek-TUI (新增 21.8K stars) 专为 DeepSeek 模型打造的终端编程智能体,直接在你的终端里运行 仓库地址: 2. UI-TARS-desktop (新增 3.2K stars) 开源多模态 AI 智能体技术栈:连接最前沿的 AI 模型与 Agent 基础设施 仓库地址: 3. CloakBrowser (新增…
@pritipatelfgoo · x.com · 1 day ago
处理超长专业文档时,向量检索常常找不准关键段落,因为它只看相似度、不看文档结构。 VectifyAI/PageIndex 换了一条路:用 LLM 推理替代向量数据库,通过文档的自然层次结构(章节、页码)进行检索,模仿人类专家浏览复杂文档的方式。 GitHub: 主要功能: 将文档组织成层次化树状索引,而非人工切块 根据完整对话上下文和领域知识自适应检索…
@rwayne · x.com · 2 days ago
6/几个值得注意的点 树索引构建有成本。 第一次处理文档时需要调用 LLM 生成树结构,一份 100 页的 PDF 大概消耗 1-2 美元,但索引建好之后可以反复用,后续每次查询的成本就很低了,如果你的文档不经常变动,这个一次性成本完全可以接受。 支持 Vision 模式。 不用 OCR,直接把页面当图片喊给模型看,这对扫描件、表格密集型文档、复杂排版的 PDF 特别有用,传统 OCR…
@sitinme · x.com · 2 days ago
These are the BEST GitHub repos this week! Bookmark this post right now: 1. DeepSeek-TUI (+21.8K stars) Coding agent for DeepSeek models that runs in your terminal Repo → 2. UI-TARS-desktop (+3.2K…
@sharbel · x.com · 3 days ago
Investing early and persisting everything you do in claude code to docs comes in handy for things you initially don't necessarily think about. E.x. I was doing a landing page re-write and after CC…
@tarek_kekhia · x.com · 3 days ago
Repository activities
repository's daily and monthly activities across stars, forks, merged PRs, issues, and closed issues
GitHub trending history
Shows when the repository has appeared on GitHub Trending across any language
all language ranking
python ranking