requestsession

requestsession / IndexGPT

Public

IndexGPT: 本地 AI 科研助手（RAG + SFT + LoRA）| Local AI research assistant for PDF parsing, retrieval QA, SFT data generation, and LoRA training.

requestsession.github.ioIndexGPT-site
73
1
100% credibility
Found Feb 12, 2026 at 1 stars 163x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

IndexGPT is a local web app for uploading research papers, searching them intelligently, chatting with evidence-based answers, and optionally training a custom AI assistant on your documents.

How It Works

1
🖥️ Discover IndexGPT

You find this handy research helper online and download it to your computer to organize your papers.

2
🧠 Add smart thinking tools

You place the AI's thinking folders in a special spot and pick them in settings so it can understand language.

3
🚀 Open the friendly web app

You start the program and a colorful dashboard appears on your screen ready for action.

4
📚 Upload your research papers

You drag in your PDF files and they land safely in your personal library.

5
🔍 Build the paper knowledge base

You click to create a smart map of all your papers so questions find the right spots instantly.

6
💬 Chat and get perfect answers

You ask about your papers and get clear replies with direct quotes from the exact pages.

7
🏋️ Teach it even more

You run lessons from your papers to make your helper smarter just for your work.

🎉 Your personal research buddy

Now you have a trusty sidekick that knows your papers inside out and speeds up your discoveries.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 1 to 73 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is IndexGPT?

IndexGPT delivers a local research assistant for PDF parsing, retrieval QA, SFT data generation, and LoRA training in one web app. Upload papers to auto-build RAG indexes, chat with evidence-cited answers linking back to source pages, then one-click generate training data and fine-tune adapters—all offline. Python backend with JavaScript frontend (Vite/React) handles it via simple tabs: Documents, AI Assistant, Training Lab, Settings.

Why is it gaining traction?

It collapses the messy RAG-to-LoRA pipeline into a no-fuss UI, where answers always cite PDF snippets to kill hallucinations—rarer in scattered scripts. Local model support plus OpenAI-compatible API fallback speeds prototyping without vendor lock-in. Niche appeal for indexgpt how to use in research drives its 65 stars.

Who should use this?

Researchers knee-deep in literature reviews needing grounded QA over dozens of PDFs. ML teams at firms like JPMorgan (indexgpt jpmc, indexgpt jp morgan) or Shopify building internal assistants from reports. Anyone prototyping local retrieval agents tired of gluing parsers, vector stores, and trainers.

Verdict

Solid pick for local research workflows—bilingual docs and CLI/web parity make indexgpt login-free setup a breeze. 1.0% credibility and low stars signal early maturity, so validate on toy datasets first; still, MIT license lowers the bar to try.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.