You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
However, the server is producing very poor text for the same input file as the cli. The poor text is similar to what I was getting before using vlm. Here's the output of running the server with dev:
INFO: 2025-04-23 10:49:29,967 - docling_serve.docling_conversion - artifacts_path is unset. The system will download the model weights at runtime.INFO: Application startup complete.INFO: 2025-04-23 10:49:45,458 - docling_serve.app - Received 1 files for processing.INFO: 2025-04-23 10:49:45,458 - docling_serve.docling_conversion - artifacts_path is unset. The system will download the model weights at runtime.INFO: 2025-04-23 10:49:45,492 - docling.document_converter - Going to convert document batch...INFO: 2025-04-23 10:49:45,493 - docling.document_converter - Initializing pipeline for StandardPdfPipeline with options hash cb917a501917ffd372a2e69dae6fd392INFO: 2025-04-23 10:49:45,536 - docling.utils.accelerator_utils - Accelerator device: 'mps'INFO: 2025-04-23 10:49:46,948 - docling.utils.accelerator_utils - Accelerator device: 'mps'INFO: 2025-04-23 10:49:47,957 - docling.utils.accelerator_utils - Accelerator device: 'mps'INFO: 2025-04-23 10:49:48,121 - docling.models.factories.base_factory - Loading plugin 'docling_defaults'INFO: 2025-04-23 10:49:48,121 - docling.models.factories - Registered picture descriptions: ['vlm', 'api']INFO: 2025-04-23 10:49:48,121 - docling.pipeline.base_pipeline - Processing document inis_iaea_org-25003257.pdfINFO: 2025-04-23 10:50:12,747 - docling.document_converter - Finished converting document inis_iaea_org-25003257.pdf in 27.29 sec.INFO: 2025-04-23 10:50:12,747 - docling_serve.response_preparation - Processed 1 docs in 27.29 seconds.INFO: 127.0.0.1:64635 - "POST /v1alpha/convert/file HTTP/1.1" 200 OK
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
How do I duplicate the functionality of the docling cli using the server? Specifically, I have found that the following works well for my documents:
I've tried the following parameters following the python usage examples https://github.com/docling-project/docling-serve/blob/main/docs/usage.md:
However, the server is producing very poor text for the same input file as the cli. The poor text is similar to what I was getting before using vlm. Here's the output of running the server with dev:
What am I doing wrong?
Beta Was this translation helpful? Give feedback.
All reactions