Docling IBM models

AI modules to support the Docling PDF document conversion project.

TableFormer is an AI module that recognizes the structure of a table and the bounding boxes of the table content.

Install

The package provides two variants which allow to seemlessly switch between opencv-python and opencv-python-headless.

# Option 1: with opencv-python-headless
pip install "docling-ibm-models[opencv-python-headless]"

# Option 2: with opencv-python
pip install "docling-ibm-models[opencv-python]"

Pipeline Overview

Architecture

Datasets

Below we list datasets used with their description, source, and "TableFormer Format". The TableFormer Format is our processed version of the version of the original format to work with the dataloader out of the box, and to augment the dataset when necassary to add missing groundtruth (bounding boxes for empty cells).

Name	Description	URL
PubTabNet	PubTabNet contains heterogeneous tables in both image and HTML format, 516k+ tables in the PubMed Central Open Access Subset	PubTabNet
FinTabNet	A dataset for Financial Report Tables with corresponding ground truth location and structure. 112k+ tables included.	FinTabNet
TableBank	TableBank is a new image-based table detection and recognition dataset built with novel weak supervision from Word and Latex documents on the internet, contains 417K high-quality labeled tables.	TableBank

Models

TableModel04:

TableModel04rs (OTSL) is our SOTA method that using transformers in order to predict table structure and bounding box.

Configuration file

Example configuration can be found inside test tests/test_tf_predictor.py These are the main sections of the configuration file:

dataset: The directory for prepared data and the parameters used during the data loading.
model: The type, name and hyperparameters of the model. Also the directory to save/load the trained checkpoint files.
train: Parameters for the training of the model.
predict: Parameters for the evaluation of the model.
dataset_wordmap: Very important part that contains token maps.

Model weights

You can download the model weights and config files from the links:

TableFormer Checkpoint

Inference Tests

You can run the inference tests for the models with:

python -m pytest tests/

This will also generate prediction and matching visualizations that can be found here: tests\test_data\viz\

Visualization outlines:

Light Pink: border of recognized table
Grey: OCR cells
Green: prediction bboxes
Red: OCR cells matched with prediction
Blue: Post processed, match
Bold Blue: column header
Bold Magenta: row header
Bold Brown: section row (if table have one)

Metadata

Release files for docling-ibm-models 4.0.3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for docling-ibm-models 4.0.3
File	Size	Uploaded
docling_ibm_models-4.0.3.tar.gz	77.6 kB	Sep 18, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for docling-ibm-models 4.0.3
File	Interpreter	ABI	Platform	Reset
docling_ibm_models-4.0.3-py3-none-any.whl 69.4 kB Sep 18, 2026	Python 3	none	any	Details

Total release size: 147.1 kB

Release files / docling_ibm_models-4.0.3.tar.gz

Download URL	docling_ibm_models-4.0.3.tar.gz
Size	77.6 kB
Tags	Source
SHA-256 checksum How to use checksums	`509e89af75e06b48500977dca11ecb6022e81b1526c90dd54587a60ccc82bdb6`
BLAKE2b-256 checksum How to use checksums	`69ee7dd19487bf333320f0fc73eddcb42347d9f266310ebb1a605f120987f7c1`
Upload date	Sep 18, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/docling-project/docling-ibm-models
Publishing commit: github.com/docling-project/docling-ibm-models/tree/759c41a34c1c5148dc66901ef97d8962f445a152

Workflow

Publishing configuration: github.com/docling-project/docling-ibm-models/blob/759c41a34c1c5148dc66901ef97d8962f445a152/.github/workflows/pypi.yml
Publishing logs: github.com/docling-project/docling-ibm-models/actions/runs/35367288343/attempts/1

Artifact

Subject: docling_ibm_models-4.0.3.tar.gz
SHA-256 checksum: 509e89af75e06b48500977dca11ecb6022e81b1526c90dd54587a60ccc82bdb6 Verifying attestations

Release files / docling_ibm_models-4.0.3-py3-none-any.whl

Download URL	docling_ibm_models-4.0.3-py3-none-any.whl
Size	69.4 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`82c6c6ad6622aa9e58c4d745c26fba12eef54927f32d44f21f24bcb5be5464ff`
BLAKE2b-256 checksum How to use checksums	`7aab395243c3825803044c58ef28131f50ff3db366c72b0f73c5c190f7b02f81`
Upload date	Sep 18, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Provenance

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/docling-project/docling-ibm-models
Publishing commit: github.com/docling-project/docling-ibm-models/tree/759c41a34c1c5148dc66901ef97d8962f445a152

Workflow

Publishing configuration: github.com/docling-project/docling-ibm-models/blob/759c41a34c1c5148dc66901ef97d8962f445a152/.github/workflows/pypi.yml
Publishing logs: github.com/docling-project/docling-ibm-models/actions/runs/35367288343/attempts/1

Artifact

Subject: docling_ibm_models-4.0.3-py3-none-any.whl
SHA-256 checksum: 82c6c6ad6622aa9e58c4d745c26fba12eef54927f32d44f21f24bcb5be5464ff Verifying attestations