Ray

Hi Vineet, Thank you for your thoughtful questions and for your interest in PageIndex! We are sorry for not getting back to you sooner: we are currently a bit short...

> Hi, > > Great efforts guys 💯 > > Let's say we have an exiting application that uses LlamaIndex or Langchain, do you have a vision on how to...

Hi, thanks for raising the issue. The example uses the PageIndex SDK (API) for generating PageIndex trees. `utils.create_node_mapping()` is part of our pip package: ``` pip install pageindex ``` Once...

Thanks for raising this! We just double-checked and can indeed do `from pageindex import PageIndexClient` without errors. A few things that might explain why you are seeing `cannot import name...

Hi, thanks for reporting this and sorry for the delayed response. We’re currently a bit short on manpower and catching up on issues. From the traceback you shared, the error...

> I had previously resolved the issue, but after pulling the latest code, I encountered an error when running the following command: > > `python3 run_pageindex.py --pdf_path '/Users/skype/Documents/GitHub/PageIndex/docs/2023-annual-report.pdf'` > >...

Good point. Cost tracking would be useful. We are considering exposing simple usage metrics (e.g., tokens processed per query / per document) so that teams can plug them into their...

Yes! You can run PageIndex with local models — e.g., via Ollama or Hugging Face — no paid API needed.

The output isn’t wrong. `start_index` and `end_index` are both **inclusive** page numbers. By default it only returns summaries. If you want text content, use `--add-node-text` flag, or simply read from...

Sure. See below for instructions to use our open-source repo: https://github.com/VectifyAI/PageIndex/?tab=readme-ov-file#-package-usage In our cookbooks, you can simply replace the tree-generation part with the open-source version; the rest of the code...