Repository navigation
extract text from PDF documents for models that don't support PDF input? #14270
Description
Question
no models from opencode zen support PDF input. when i attempt to read PDF documents using Kimi K2.5, Big Pickle or MiniMax M2.5, the tool call returns "PDF read successfully" but the model fails to process the document (see screenshots). would it be possible to extend the read tool to extract text from the document and give it to the model?
attempting to read a PDF document with Kimi K2.5
same with MiniMax 2.5 Free
Activity
github-actions commentedon Feb 19, 2026
This issue might be a duplicate of existing issues. Please check:
readtool unable to handle PDFs and suggestion to add text extraction support for models that don't natively support PDF inputIf you believe your issue is distinct, please clarify what differentiates it from the above.
agnlt64 commentedon Feb 19, 2026
the above issue (#3383) is about a bug in the
readtool while trying to process PDF documents. it would crash with this error message:AI_UnsupportedFunctionalityError: 'file part media type application/pdf' functionality not supported. in my case, the tool doesn't crash and says that the document was read successfully. there is no bug, i just want to know if it's possible to extend thereadtool to extract text from PDF documents and send it to models that don't support PDF as input.github-actions commentedon Apr 21, 2026
To stay organized issues are automatically closed after 60 days of no activity. If the issue is still relevant please open a new one.