Blame for README.md

---
license: other
pretty_name: WholeDoc Bench (sample)
task_categories:
- image-to-text
language:
- "de"
- "en"
- "es"
- "fr"
- "nl"
size_categories:
- n<1K
tags:
- document-parsing
- pdf
- benchmark
- evaluation
- tables
- charts
- ocr
- layout-detection
- barcode
- mrz
configs:
- config_name: default
  data_files:
  - split: test
    path: documents.jsonl
citation: |
  @misc{wholedocbench2026,
    title  = {WholeDoc Bench: Continuity, Fidelity and Machine-Readable Content in Document Parsing},
    author = {Moldovan, Denis and Sookun, Shrevin and Mogage, Nicolae and Pizur, Matus and Mitrea, Costinel and Houssin, Olivier and Alby, Etienne and Carlier, Coralie and Mart{\'\i}n, Daniel},
    year   = {2026},
    url    = {https://github.com/PSPDFKit-labs/wholedoc-bench}
  }
---

# WholeDoc Bench

**Quick links:** [\[🌐 Leaderboard\]](https://www.nutrient.io/benchmarks/parse/) [\[📜 Paper\]](https://github.com/PSPDFKit-labs/wholedoc-bench/blob/main/paper.pdf) [\[💻 Code\]](https://github.com/PSPDFKit-labs/wholedoc-bench)

This is the **public sample** of WholeDoc Bench, a benchmark for evaluating document parsers on
whole documents. It contains 10 documents (28 pages) with their ground-truth answer keys, so you
can inspect the annotation format and run the evaluation pipeline on these documents.

Results from the full private evaluation set are available on the
[benchmark page](https://www.nutrient.io/benchmarks/parse/) and in the
[paper](https://github.com/PSPDFKit-labs/wholedoc-bench/blob/main/paper.pdf). Those results
cannot be reproduced from this sample alone. The
[evaluation code](https://github.com/PSPDFKit-labs/wholedoc-bench) is available to run and score
supported parsers.

## Usage

Download the whole release, pinned to its tag. The rows alone are not enough to score:

```bash
uvx --from huggingface_hub hf download nutrientdocs/wholedoc-bench-sample --repo-type dataset --revision v1 --local-dir wholedoc-bench-sample
```

Then run a parser, score it and build a leaderboard with the [evaluation code](https://github.com/PSPDFKit-labs/wholedoc-bench):

```bash
uv run wholedoc-bench run --provider pdftotext --corpus wholedoc-bench-sample --out runs/pdftotext
uv run wholedoc-bench score --corpus wholedoc-bench-sample --predictions runs/pdftotext --results results
uv run wholedoc-bench report --results results
```

Every command checks the release first and stops if a file is missing or changed.

<details>
  <summary>Files and row fields</summary>

```
documents.jsonl       one row per document
documents/<id>.pdf    the documents
groundtruth/<id>.json the ground truth of each document
release.json          the release tag, its source commit and counts
SHA256SUMS            the checksum of every file above
published.jsonl       every document id ever published, with its first release
```

| Field | Meaning |
|---|---|
| `id` | the document's permanent id |
| `document_path`, `document_sha256` | the document file and its checksum |
| `format`, `pages` | file format and page count |
| `language`, `scan` | languages, and whether the document is a scan |
| `tracks`, `supported_metrics` | what this document is scored on |
| `groundtruth_source` | how the ground truth was made |
| `groundtruth_path`, `groundtruth_sha256` | the ground truth file and its checksum |
| `category`, `limits`, `licence`, `attribution` | not recorded in this release |

</details>

## Public sample and private set

The published benchmark results use the full private evaluation set. This public sample lets you
inspect the answer keys and run the evaluation pipeline, but it does not reproduce the full
benchmark results.

## Citation

```bibtex
@misc{wholedocbench2026,
  title  = {WholeDoc Bench: Continuity, Fidelity and Machine-Readable Content in Document Parsing},
  author = {Moldovan, Denis and Sookun, Shrevin and Mogage, Nicolae and Pizur, Matus and Mitrea, Costinel and Houssin, Olivier and Alby, Etienne and Carlier, Coralie and Mart{\'\i}n, Daniel},
  year   = {2026},
  url    = {https://github.com/PSPDFKit-labs/wholedoc-bench}
}
```

## About the author

<a href="https://nutrient.io/">
  <img src="https://avatars2.githubusercontent.com/u/1527679?v=3&s=200" height="80" />
</a>

This project is maintained and funded by [Nutrient](https://nutrient.io/) - The deterministic document infrastructure enterprises run their highest-stakes workflows on: replayable output, clear exceptions, and full audit trails on the messy, regulated documents where AI alone breaks.