PDF Memory
Overview
At the moment, PDF ingestion memories do not support images inside the PDF but only text content.
PDF ingestion is limited to 1000 pages and up to 50 MB PDF files.
Hana system does not store your actual PDF files. We simply process your PDF at runtime, extract text content from it and embed and ingest it inside the system but the actual single PDF file itself is not stored. So, it is not possible to download the PDF file from Hana dashboard later on.
- Hana will ingest PDFs and will seamlessly consider their context in while generating a response.
Accessing PDF Memory
-
Navigate to the Memory section from the left sidebar.
-
Click on the PDF Memory tab to view all PDF memories.
Adding a PDF Memory
- Click the drag and drop area to open file selector or drag and drop a PDF file into this area to start ingesting.
The PDF will immediately start ingesting, so choose the file carefully. If you have ingested the wrong file, you can delete it once it has finished processing.
- The ingestion status will initially show as IN_PROGRESS, which updates to INGESTED once the processing is complete.
Deleting PDF Memory Entries
-
Use the corresponding Delete option for a specific entry.
-
Deleting removes the selected PDF memory batch and all its content from Hana's memory.