The LLM Demo Index: 25 Projects and Not a Monorepo in Sight
To save developers from hunting for LLM starter templates, this repo catalogs 25 independent projects—RAG, chatbots, fine-tuning, and more—mostly using LangChain, Gemini, and Streamlit.

What it does
This repository is essentially a curated syllabus: a single README table linking to 25 separate Generative AI projects. Each entry is a standalone demo or tutorial—PDF chatbots, YouTube summarizers, text-to-SQL converters, medical RAG systems, and fine-tuning recipes—mostly built with Python, LangChain, and Streamlit. It does not contain the code itself; it maps out where to find it.
The interesting bit
The collection has a noticeable medical and paperwork tilt, with multiple entries for medical chatbots, invoice extraction, and resume parsing. It also juxtaposes proprietary APIs like Google Gemini Pro and Groq against fully open-source stacks such as Mistral 7B, BioMistral, and Ollama, letting you compare the same architectural patterns with and without a cloud bill.
Key highlights
- Covers a wide spread: RAG with LlamaIndex, text-to-SQL, on-device inference via MediaPipe, Stable Diffusion speedups, and even a CrewAI financial-analysis agent.
- Fine-tuning is represented through LoRA, QLoRA, Unsloth, and DPO training recipes for Llama 2 and Phi-2.
- Several projects target fully open-source deployment: Haystack + Mistral 7B, Gemma-7B via Ollama, and BioMistral + Qdrant + PubMedBERT.
- Most demos deploy through Streamlit or Gradio, with a few using FastAPI or Chainlit.
- The table links directly to each project’s own repository, making this a meta-directory rather than a monorepo.
Caveats
- This is strictly an index; code and documentation quality will vary across the 25 linked repositories.
- Several table entries have no description at all beyond a project name, so you may need to click through to understand what a demo actually does.
- The README offers no filtering, tagging, or difficulty ratings, which makes browsing feel like scrolling through an unsearchable spreadsheet.
Verdict
Worth bookmarking if you are a developer looking for starter templates or comparing how Gemini Pro stacks up against open-source alternatives in real mini-apps. Skip it if you want a single maintained library or framework—this is a directory, not a product.
Frequently asked
- What is GURPREETKAURJETHRA/END-TO-END-GENERATIVE-AI-PROJECTS?
- To save developers from hunting for LLM starter templates, this repo catalogs 25 independent projects—RAG, chatbots, fine-tuning, and more—mostly using LangChain, Gemini, and Streamlit.
- Is END-TO-END-GENERATIVE-AI-PROJECTS open source?
- Yes — GURPREETKAURJETHRA/END-TO-END-GENERATIVE-AI-PROJECTS is open source, released under the MIT license.
- How popular is END-TO-END-GENERATIVE-AI-PROJECTS?
- GURPREETKAURJETHRA/END-TO-END-GENERATIVE-AI-PROJECTS has 610 stars on GitHub.
- Where can I find END-TO-END-GENERATIVE-AI-PROJECTS?
- GURPREETKAURJETHRA/END-TO-END-GENERATIVE-AI-PROJECTS is on GitHub at https://github.com/GURPREETKAURJETHRA/END-TO-END-GENERATIVE-AI-PROJECTS.