FAQ
Frequently asked questions about LlamaParse, covering topics like credit usage, document storage, parsing time, and current limitations.
Pricing & Credits
Section titled “Pricing & Credits”How do credits work?
All features are priced using credits, which are billed per page (or minute for audio). Credits vary by parsing mode, model, and whether files are cached. Check the Credit Pricing & Usage page for full details.
Privacy & Caching
Section titled “Privacy & Caching”Are the documents stored or cached anywhere, if so for how long?
Privacy is an important consideration when parsing sensitive documents. Your data is kept private to you only and is used only to return your results, never for model training. To avoid charging multiple times for parsing the same document, your files are cached for 48 hours and then permanently deleted from our servers.
How to avoid caching sensitive documents?
If you wish to avoid caching sensitive documents, you may set do_not_cache=True.
Parsing
Section titled “Parsing”How long does parsing typically take?
It depends on the length and complexity of the document, in particular:
- The number of pages
- The number of images
- And whether text must be extracted from those images
- The density of text on each page
On average, parsing a block of pages takes 45 seconds. However, this is a rough estimate and the actual time may vary. For example, a document with many images may take longer to parse than a text-only document with the same number of pages.
Limits
Section titled “Limits”What are some of the current limitations?
See the Limitations page for service-specific limits on file size, page counts, schema constraints, timeouts, and more.
Support
Section titled “Support”How do I get help or report an issue?
If you are experiencing issues with LlamaParse, check the Status Page for ongoing incidents. You can also review the Troubleshooting & Error Codes guide. If you are being throttled, review the Rate Limits page. For unresolved issues, contact support@runllama.ai.