Verified purchaser
I use as a more powerful NotebookLM
I had a project where i needed to upload a lot of scanned historic documents. I have about 200 pdfs and about 4000 pages.
Firstly, the OCR on this is really good. These docs dates from the 1940s-90s. They are badly scanned and annotated but the OCR deals with it.
The labelling system gets around the context limits of LLMs as best as i think is currently possible.
It would be nice to be able to save notes / extractions to the memory of the folder so you create a folder-wide context without having to include all of the docs.
I opted for Tier 2. All of this OCR work and analysis and i still have monthly credits remaining.
Bravo, team.
Christophe_docAnalyzer.ai
May 1, 2025Thanks so much for the thoughtful feedback—and we're glad to hear docAnalyzer is handling those challenging historical documents well!
You're absolutely right: your idea about notes is very aligned with what we're building next. In fact, enhancing notes and enabling persistent, workspace-level memory is a key part of our upcoming roadmap—expect to see major updates on that front this summer.