surya · Issues· 198 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #555
Nothing happens when I start OCR
Updated Sep 16, 2026 - #173
Finetuning for layout detection
Updated Sep 2, 2026 - #536
[Bug: Breaking] `AttributeError: 'PdfDocument' object has no attribute 'render'` in Streamlit app
bug: breakingUpdated Aug 23, 2026 - #540
RecognitionPredictor deterministically outputs Cyrillic/Armenian homoglyphs for specific short German text lines (llamacpp backend)
Updated Aug 20, 2026 - #544
[BUG: Output] Text inside Diagram/Figure regions is never OCR'd — full_page returns an empty skipped block (detection+recognition reads it fine)
Updated Aug 19, 2026 - #551
Full-page OCR narrates a table as a <p> caption instead of transcribing it — every figure is lost (surya-ocr-2)
Updated Aug 19, 2026 - #549
[FEAT] Support podman
enhancementUpdated Aug 19, 2026 - #480
Update Pillow version constraint to support Python 3.14
Updated Aug 17, 2026 - #547
vLLM container is published on all interfaces (-p {port}:8000 ignores SURYA_INFERENCE_HOST)
Updated Aug 12, 2026 - #543
[BUG: Breaking] vLLM EngineCore crashes with SIGSEGV (`realloc(): invalid pointer` / `EngineCore: -11`) during startup when loading `datalab-to/surya-ocr-2`
bug: breakingUpdated Jul 28, 2026 - #542
[BUG: Breaking] Inference error with llama.cpp backend: failed to parse grammar due to \d escape in bbox schema
bug: breakingUpdated Jul 27, 2026 - #541
How to reproduce the 83.3% olmOCR-bench score? What are the "adjustments applied to account for our output HTML format"?
bug: outputUpdated Jul 24, 2026 - #538
Verify evals on Papers with Code
Updated Jul 22, 2026 - #523
[BUG: Breaking] Parallel inference not working on llama.cpp backend.
bug: breakingUpdated Jul 21, 2026 - #531
[BUG: Breaking] Import Error
bug: breakingUpdated Jul 13, 2026