Guides for document-to-text pipelines.
Practical write-ups on extracting, OCR-ing, and parsing documents for RAG, search, and LLM ingestion — no fluff, just how to actually do it.
- Jul 14, 2026Per-document vs per-page pricing for document extractionPer-page pricing sounds cheaper until you multiply it by a 300-page report. Here's the framing that actually predicts your bill, and the case where per-page really does win.
- Jun 23, 2026Ingesting large documents and big batches without falling overA 500-page PDF and a backfill of ten thousand files both stress an ingestion pipeline in the same two places: single-request time, and how many requests you fire at once. Here's how to handle both against a synchronous endpoint.
- Jun 2, 2026Extracting tables from PDFs and spreadsheets for RAGTables don't survive naive text extraction — they collapse into number-soup with no row or column structure. Here's how to get tabular content into a shape your chunker can actually work with.
- May 5, 2026Using txtfetch as a LangChain and LlamaIndex document loaderThere's no official txtfetch SDK yet, but wiring the HTTP endpoint into LangChain's document loader interface or a LlamaIndex reader is about fifteen lines. Here's both.
- Apr 7, 2026Chunking strategies for RAG: from clean text to good retrievalFixed-size, recursive, and structure-aware chunking all assume you're starting from clean extracted text. Here's how the three compare, and why extraction quality bounds chunk quality no matter which one you pick.
- Mar 10, 2026Parsing DOCX, PPTX, and XLSX into clean text for LLM ingestionpython-docx, python-pptx, and openpyxl each have their own API, their own edge cases, and their own way of failing on real-world files. Here's a single-endpoint alternative.
- Feb 3, 2026OCR scanned documents and images through one API callScanned invoices, faxed contracts, photographed receipts — none of them have a text layer. Here's how to OCR them through the same endpoint you use for everything else.
- Jan 12, 2026How to extract text from a PDF for RAG (without maintaining a parser)Feeding PDFs into a RAG pipeline breaks the usual parser stack — multi-column layouts, embedded tables, scanned pages. Here's a one-request approach that handles all three.
Stop parsing. Start shipping.
Create an account and get an API key in minutes — the free Hobby plan needs no card.
Get started →