Self-Hosted
Free
Free self-hosted open-source tier on the pricing page.
- Deployment
- Deploy on your own infrastructure
- License
- Open source on GitHub
- Support
- Community support
Images · Vendor site 2026-10-04
thepipe (The Pipe) is an open-source document AI pipeline and hosted API that extracts clean markdown or structured data from PDFs, slides, spreadsheets, and scans. Vision-language models locate tables and figures, support Pydantic schemas, chunking for RAG, and optional on-prem deployments.
Teams building RAG or agent pipelines that need reliable parsing of messy PDFs and office documents without maintaining custom OCR stacks.
Hobby hosted tier includes one million tokens per month and 20 MB files; Scale raises limits and Business requires a sales chat for on-prem support.
thepipe markets an AI-native ETL layer that turns tricky documents into clean markdown or structured records for downstream LLM and RAG systems.
What it can do
Unknown is different from unavailable. Each fact carries its own evidence.
| Capability | Value | Evidence | Checked |
|---|---|---|---|
| Extraction | Marketing promises markdown, tables, figures, and equations even from scanned PDFs without a text layer. | Facts sourced | 2026-10-04 |
| Structured output | Pipelines can return JSON or CSV conforming to user Pydantic schemas via vision-language models. | Facts sourced | 2026-10-04 |
| Hobby pricing | Pricing lists Hobby at $25 per month with one million tokens included. | Facts sourced | 2026-10-04 |
Understand the total cost
Self-Hosted
Free
Free self-hosted open-source tier on the pricing page.
Hobby
$25.00 /mo
$25 per month Hobby hosted plan on thepipe pricing.
Article Factory is a document-centric AI writing app for blogs, news, and long-form content with paired photo generation. You define topic, language, and tone, then invoke models such as GPT-5, Claude, or Gemini to draft sections you can edit, regenerate, and export.
Explore toolReceiptUp provides a receipt OCR API and subscription scans for extracting structured data from receipt images with live and queued processing endpoints documented for developers.
Explore toolStablecog is a hosted Stable Diffusion image generator with history, gallery, and paid credit plans. Subscribers get faster parallel generations, private images, and optional commercial use depending on tier.
Explore toolThe practical questions
thepipe (The Pipe) is an open-source document AI pipeline and hosted API that extracts clean markdown or structured data from PDFs, slides, spreadsheets, and scans. Vision-language models locate tables and figures, support Pydantic schemas, chunking for RAG, and optional on-prem deployments.
Self-hosted open-source deployment is free on GitHub with bring-your-own keys and community support.. This record lists ongoing free access; check the plan limits before starting.
The listed monthly price is $25.00 USD at monthly billing. See the plan cards for entitlements, billing commitments and seat minimums.
Developer docs and hosted plans expose scrape and extraction endpoints with GPU-accelerated cloud or local deployment options.. API access and subscription access may have different terms; consult the linked sources.
Hobby hosted tier includes one million tokens per month and 20 MB files; Scale raises limits and Business requires a sales chat for on-prem support.
No retained pricing changes yet. A current price alone does not establish a historical trend.
Reviewed vendor source
Read original source ↗Facts apply to the named version and check date. Send a sourced correction if something changed.