Dataset Viewer
Auto-converted to Parquet Duplicate
pdf
unknown
doc_id
stringclasses
1 value
markdown
stringclasses
1 value
surya_blocks
stringclasses
1 value
inference_info
stringclasses
1 value
"JVBERi0xLjcKJb/3ov4KMSAwIG9iago8PCAvTWV0YWRhdGEgMyAwIFIgL05hbWVzIDQgMCBSIC9PcGVuQWN0aW9uIDUgMCBSIC9(...TRUNCATED)
arxiv-2606.17162v1
"arXiv:2606.17162v1 [cs.CL] 15 Jun 2026\nMemSlides: A Hierarchical Memory Driven Agent Framework for(...TRUNCATED)
"[{\"blocks\": [{\"polygon\": [[40.800000000000004, 589.248], [96.288, 589.248], [96.288, 1514.30399(...TRUNCATED)
"[{\"model\": \"datalab-to/surya-ocr-2\", \"model_name\": \"surya-ocr-2\", \"column_name\": \"markdo(...TRUNCATED)

Surya OCR 2 (ocr) on davanstrien/surya-smoke-pdf-src

Full-page ocr (structured html + bounding boxes) over PDF documents in davanstrien/surya-smoke-pdf-src using Surya OCR 2 (650M, Qwen3.5-based) by Datalab, via the surya-ocr package, run as offline vLLM batch inference on Hugging Face Jobs.

Processing Details

  • Source Dataset: davanstrien/surya-smoke-pdf-src
  • Model: datalab-to/surya-ocr-2
  • Task: ocr
  • Input column: pdf (PDF, pages 0-2)
  • Text column: markdown (flattened, reading-order text per row)
  • Structured column: surya_blocks (JSON: per-page blocks with bbox / polygon / label / reading_order / confidence / html)
  • Split: train
  • Samples: 1
  • Processed OK: 1 / 1
  • Processing time: 3.7 min
  • Date: 2026-06-22 13:50 UTC

License note

Surya's code is Apache-2.0, but the model weights use a modified OpenRAIL-M license: free for research, personal use, and startups under $5M funding/revenue, restricted from competitive use against Datalab's API. See the model card.

Dataset Structure

Original columns plus:

  • markdown: flattened text (OCR), label outline (layout), or table HTML (table)
  • surya_blocks: structured result as a JSON string (one entry per page)
  • inference_info: JSON list tracking models applied to this dataset

Generated with UV Scripts.

Downloads last month
33