Getting Started

Seqtra combines chunking and retrieving in one stage, i.e. chunking only with respect to the query during retrieval rather than defining chunk boundary during the ingestion phase. This strategy is known as late chunking in literature. Please note that we only accept PDFs for now.

Setup

Python Version: Python 3.10.16

Clone the repository:

    git clone https://github.com/seqtra/SeqtraClient.git

Create an environment and do:

    pip install -r requirements.txt

Test on your own input

Get your own API Key

To use Seqtra's services, you'll need to obtain a free API key:

1. Visit app.seqtra.com

2. Create an account or log in to your existing account

3. Navigate to the API section in your dashboard

4. Generate a new API key

5. Copy the key and add it to the api_token field in the streamlit app

Your API key provides secure access to Seqtra's advanced document processing capabilities at no cost.

Using UI

Run the following:

    streamlit run client_ui.py