OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

IndianTaxGPT

IndianTaxGPT is a Python question-answering application that retrieves passages from user-provided Indian income-tax PDFs and generates plain-language answers with source-file and page citations. It combines a Streamlit chat interface, LangChain retrieval, MiniLM embeddings, a Pinecone serverless vector index and a local Llama 2 7B GGUF model.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

IndianTaxGPT — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you have Indian income-tax PDFs to index and want a cited, plain-language way to locate relevant provisions or guidance. Its answers are grounded in the indexed document set, so the PDFs need to represent the law or period you want to research.

Who it is for

  • Individuals and freelancers looking up tax rules before filing
  • Small-business owners checking routine questions such as advance-tax timing
  • Accountants and CA firms researching their own collection of Acts, rules and circulars
  • Students studying Indian income-tax provisions

Documented uses

  • Find relevant passages in a collection of Indian tax PDFs and see their source pages.
  • Look up where a rule about a deduction, tax form or payment timing is addressed.
  • Search an organization's indexed tax documents without sending question text to an AI provider.
  • Refresh the assistant's reference material by replacing PDFs and re-running ingestion.
Before you choose

Requirements & limits

The repository is public and identified as MIT-licensed. Running the application requires a Pinecone account and API key; the README describes a free Pinecone plan for a typical document set.

  • The default model requires about 8 GB RAM, 5 GB disk space and a download of roughly 4.1 GB.
  • The language model and embeddings run locally, but the Pinecone vector index is hosted; the project says only the question embedding, not the question text, is sent to the vector database.
  • Answer coverage depends on which documents have been added and indexed.
  • The documented stack includes Python 3.11+, Llama 2 7B GGUF via ctransformers, all-MiniLM-L6-v2, Pinecone serverless, LangChain and Streamlit.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up IndianTaxGPT and index tax PDFs

Install the application, configure its Pinecone connection, download the model and build a searchable index from tax PDFs.

Before you start

  • Python 3.11 or newer
  • About 8 GB RAM and 5 GB disk space for the default model
  • A Pinecone account and API key
  • Tax PDFs to place in the data directory

Steps

  1. Clone the repository, enter its directory, create and activate a virtual environment, then install the project with the command below.

    Documented command
    pip install -e .

  2. Copy .env.example to .env and set PINECONE_API_KEY.

  3. Download the default model with the command below.

    Documented command
    python -m indiantaxgpt.download_model

  4. Place tax PDFs in data/, run the command below to check the files, then run the command below to build the index.

    Documented command
    python -m indiantaxgpt.ingest --dry-run
    Documented command
    python -m indiantaxgpt.ingest

Expected outcome

Tax PDF passages are embedded and uploaded to Pinecone, ready for the documented Streamlit chat stage.

Guide limitations

  • The setup uses a local model and embedding model but requires a Pinecone account and API key for the vector index.
  • The guide covers installation and indexing; it does not claim that these steps were independently executed.
Documentation 1 ↗Documentation 2 ↗
COMPARE OPTIONS

Compare options for Vector search

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
IndianTaxGPTFind relevant passages in a collection of Indian tax PDFs and see their source pages.Cloud, LinuxThe repository is public and identified as MIT-licensed. Running the application requires a Pinecone account and API key; the README describes a free Pinecone plan for a typical document set.
Chatbot with PDF for Semantic SearchPrepare document content as embeddings and save it to a supported vector database.WebThe repository is identified as open source and publicly visible on GitHub.
AI LLM PlaygroundAsk questions about PDF documents through an assistant interface.Web
Evidence and transparency

Sources & review

Official repository ↗Guide documentation ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
NitinReddy-A/IncomeTAXGPT · Official homepage title · retained in its source language
How we review sourcesSuggest a correction