Add -s -- plan to print every step and run nothing, verify to only check, rollback --yes to undo. Not run: commands checked against the official docs only. Commands checked against the official docs on 2026-10-15. Not run end to end.
AGENT VIEW · text/markdown · or send Accept: text/markdown
---
schema: "https://insidethemachine.ai/schema/listing.schema.json"
slug: "docling"
name: "Docling"
category: "Tools / Document Ingest"
stack_layer: "memory"
homepage: "https://github.com/docling-project/docling"
repo: "https://github.com/docling-project/docling"
docs: "https://docling-project.github.io/docling/"
license: "MIT"
platforms: ["macos-apple-silicon", "linux", "windows"]
mac_studio_fit: "Primary"
stars: 68929
stars_as_of: "2026-10-15"
last_push: "2026-10-15"
gap_score: null
gap_score_status: "pending test on Mac Studio"
benchmarks_status: "pending test on Mac Studio"
verified_on: "2026-10-15"
sponsored: false
---
# Docling
> Turn PDFs, Office docs, Apple iWork files, audio and more into clean Markdown/JSON for RAG.
**What:** Turns documents into clean text for AI. **Why:** Better input means better answers from your docs.
**Stack layer:** memory (model → runtime → harness → memory → tools)
A document conversion toolkit (Python package and CLI) that parses many formats into structured output for gen-AI pipelines.
**Gap role:** Retrieval quality: better parsing, better RAG answers from local models.
## Gap Score
null (pending test on Mac Studio)
## Prerequisites
- Python 3.10+ (3.9 support dropped in 2.70.0)
## Install
### macOS (Apple Silicon)
1. Run:
```bash
python3 -m venv ~/.venvs/docling && source ~/.venvs/docling/bin/activate
```
2. Run:
```bash
pip install docling
```
### Linux
1. Run:
```bash
pip install docling
```
### Windows
1. Run:
```bash
pip install docling
```
## Configure
1. optional: VLM pipeline using GraniteDocling:
```bash
docling --pipeline vlm --vlm-model granite_docling https://arxiv.org/pdf/2206.01062
```
## Verify
1. Expect: writes a Markdown conversion of the paper to the current directory
```bash
docling https://arxiv.org/pdf/2206.01062
```
## Rollback
1. Run:
```bash
pip uninstall -y docling && rm -rf ~/.venvs/docling
```
## One-command install
Status: docs-checked (Not run: commands checked against the official docs only). The wrapper scripts are generated from the site data. Every .sh passed bash -n and shellcheck and ran in plan mode on Linux (Debian 13). Every .ps1 passed the PowerShell parser and chat.ps1 ran in plan, verify and rollback mode under PowerShell 7 on Linux with Windows calls mocked. Not run on macOS or real Windows.
```bash
curl -fsSL https://insidethemachine.ai/install/docling.sh | bash # install, then check
curl -fsSL https://insidethemachine.ai/install/docling.sh | bash -s -- plan # print only
curl -fsSL https://insidethemachine.ai/install/docling.sh | bash -s -- rollback --yes
```
Details and Windows: https://insidethemachine.ai/install/docling.md
## Caveats
- First run downloads layout models.
## Benchmarks
Status: pending test on Mac Studio. Rig: Mac Studio, Apple Silicon, 128GB unified memory. No results published yet.
## Works with
- mem0 (https://insidethemachine.ai/setups/mem0.md)
- open-webui (https://insidethemachine.ai/setups/open-webui.md)
## Sources
- https://github.com/docling-project/docling