Autism RAG

Autism RAG

A chatbot about autism that finds passages in a stored vector database and passes them to a small language model to answer. Suits developers and students studying retrieval based chat.

Made by Shriharshan on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: This Space is a chatbot about autism built with retrieval augmented generation. A contextual retriever looks up passages in a stored vector database, and a small instruct model writes the reply from what was found. The repository holds the database folder and the app code.

Who it suits: Developers and students who want a compact retrieval chat example, and people exploring how a topic specific assistant is put together.

Industry | Who uses it | How they use it | Example
Software | RAG developers | Study a retriever plus reranker plus small model setup | Read app.py to see how the pieces connect
Education | Students of NLP | See how retrieval feeds a chat answer | Ask a question and compare the answer with the topic
Research | Information researchers | Test how a topic chatbot responds | Try a few questions on the same theme
Health information | Curious readers | Ask general questions about autism | Type a question and read the reply

If you need advice about a person's care, a chatbot is not the place to get it.

Where to use it: It runs in two places.

• A web browser, on the Space's page at Hugging Face with nothing to install
• A local copy on Windows, macOS or Linux, which needs Python

Use case | Who | What they get back
Ask a question | Readers | A written reply built from retrieved passages
Study the retriever | Developers | An example of a vector database behind a chat page
Run your own copy | Builders | The same chatbot on your own computer

How to set it up: Using it online needs nothing. A local copy needs a few things.

• Python and pip
• The packages listed in requirements.txt
• The VectorDB folder that ships with the repository

1. Install the Hugging Face tools.
2. Download the Space into a folder.
3. Move into that folder.
4. Install Gradio and the packages from requirements.txt.
5. Start the app.

pip install -U huggingface_hub
hf download Shriharshan/Autism-RAG --repo-type space --local-dir Autism-RAG
cd Autism-RAG
pip install gradio==5.23.1
pip install -r requirements.txt
python app.py

How to use it: Open the Space page and type a question about autism into the chat box. The reply appears in the same chat. In a local copy, run python app.py to get the same page on your own computer.

Example: You ask the chatbot a general question about autism. It looks up related passages in its database, then the small model writes an answer from them.

For newcomers and experts: A retrieval chatbot looks things up before it answers, which keeps replies closer to its stored material. Ask a simple question first and see how it replies.

The Space names three models. A BGE M3 model is likely the embedder, a BGE reranker v2 M3 model reorders the found passages, and a SmolLM2 360M Instruct model writes the answer. The stored vectors sit in the VectorDB folder.

You type a question in the chat
The retriever searches the vector database
A reranker orders the found passages
A small model writes the answer

Good to know: The README is a single line, so the source documents, their dates and the answer quality are not described.

• License: MIT
• Cost to run: none named
• Data it reaches: the questions you type and the files in its VectorDB folder
• Not for: advice on a person's care, which the README does not offer

On Hugging Face, 13 November 2026

• Likes: 3
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/Shriharshan/Autism-RAG
• Files: https://huggingface.co/spaces/Shriharshan/Autism-RAG/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.