Docling converts PDF and DOCX files into clean Markdown and structured JSON for search, RAG pipelines and data extraction. This Ubuntu 24.04 AMI runs Docling on CPU behind a private HTTPS API with per-instance TLS. Processing stays on the instance, with no model API key and no remote services.
Docling extracts headings, paragraphs and tables from born-digital PDF and DOCX documents and returns Markdown and structured JSON. This AMI is for developers and data teams who prepare documents for search indexes, RAG pipelines or analytics and want the conversion to run privately in their own AWS account.
What you get: Docling with local layout and table models on CPU, an HTTPS API on port 8443 with a TLS certificate and access files created on first boot, and a document conversion walkthrough under /opt/sushi/docs. No provider key is needed. Each request accepts a PDF or DOCX of up to 10 MiB and 20 PDF pages, processed one at a time on t3.large. OCR for scanned images, remote enrichment and GPU processing are not included. Review extracted content against the source document.
Getting started: launch t3.large with at least 32 GiB of encrypted gp3 storage and IMDSv2, keep inbound application ports closed and connect with Session Manager. Run sudo -iu ubuntu bash /opt/sushi/docs/demo.sh, forward port 8443 and convert your first document using the guide. Back up your data before replacing an instance.
Known issues: the 17 October 2026 host scan reports High and Critical matches in the pinned Ubuntu kernel and 72 High matches on the retained linux-libc-dev package; the Python runtime scan reports none. Read https://docs.softwaresushi.com/docling/operations/#maintenance before deployment.
This is a repackaged open source software product wherein additional charges apply for independent AWS packaging, setup documentation and maintenance by Software Sushi. Docling is published under the MIT license; model and dependency notices are kept with the installed package. Software Sushi is independent of the upstream project and does not claim upstream or AWS endorsement. EC2, EBS and data transfer are billed separately from software charges.
Highlights
Convert born-digital PDF and DOCX files into Markdown and structured JSON on CPU.
Local layout and table models: no model API key and no remote services.
Private HTTPS API with per-instance TLS, plus a guided first conversion.
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Try this product free for 5 days according to the free trial terms set by the vendor. Usage-based pricing is in effect for usage beyond the free trial terms. Your free trial gets automatically converted to a paid subscription when the trial ends, but may be canceled any time before that.
This listing uses one usage-based pricing dimension. You pay an hourly software rate for each running t3.large instance. Billing is tied to the time the instance runs, so your cost scales with how long you keep it active. There is no annual term and no bundled trial. Stopping or terminating the instance ends the hourly software charge. AWS infrastructure such as EC2, EBS, and data transfer is billed separately from this software rate.
Top-of-mind questions for buyers
What hardware do I get on the t3.large hourly rate?
The software runs on a t3.large instance with at least 8 GiB of memory and 32 GiB of root storage. It handles CPU-based conversion with local layout and table models. OCR and GPU processing are not part of this package. Access is through a private HTTPS connection on port 8443.
Am I charged when the instance is stopped or terminated?
The hourly software rate accrues only while the instance runs. Stopping or terminating the instance ends that charge. Retained EBS volumes and other AWS resources can keep incurring charges even when the instance is not running, since AWS infrastructure is billed separately from the software rate.
What file processing limits apply under this package?
You can convert PDF and DOCX files only, up to 10 MiB per file and 20 PDF pages. The tool processes one conversion at a time and needs a PDF with selectable text. Scanned pages, image or audio input, embeddings, and a GPU endpoint fall outside this package.
docs.softwaresushi.com
Helpful?
Vendor refund policy
We offer a full refund of software charges for the first 48 hours billing starts if the product does not perform as described. To request a refund, email info@softwaresushi.com with your EC2 instance ID and a description of the issue. AWS infrastructure costs (EC2, EBS, data transfer) are billed by AWS and are not eligible for refund by us.
How can we make this page better?
Tell us how we can improve this page, or report an issue with this product.
Give us feedbackReport a problem with this product or seller
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA).
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
An AMI is a virtual image that provides the information required to launch an instance. Amazon EC2 (Elastic Compute Cloud) instances are virtual servers on which you can run your applications and workloads, offering varying combinations of CPU, memory, storage, and networking resources. You can launch as many instances from as many different AMIs as you need.
Version release notes
Convert born-digital PDF and DOCX documents into Markdown and structured JSON with local CPU processing. OCR and remote services are disabled.
Additional details
Usage instructions
Launch t3.large with at least 32 GiB encrypted gp3 root storage and IMDSv2; keep inbound application ports closed. 2. Connect with AWS Systems Manager Session Manager (no key pair or inbound rule needed); you land as ssm-user. 3. Wait for cloud-init and run sudo -iu ubuntu bash /opt/sushi/docs/demo.sh. 4. Forward HTTPS port 8443 with Session Manager or an SSH tunnel; follow the private access instructions and first request in the guide at /opt/sushi/docs/README.md on the instance. 5. Back up customer data before replacing the instance. EC2, EBS and data transfer are billed separately from software.
Software Sushi provides installation and configuration guidance for this image. Contact info@softwaresushi.com with your product version and reproduction steps; never send keys or private configuration. Guaranteed response times require a separate agreement. This offer does not include upstream project support.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Store text documents and retrieve exact sources with Haystack 3.4.1 and Hayhooks 2.1.0 on Ubuntu 24.04. Add a model you supply for cited answers. This is paid repackaged open-source software.
Classify support tickets in Spanish, Portuguese, and English. Returns intent, urgency, sentiment, language, and model scores on CPU, without runtime downloads.
Be the first to review this product. We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.