Ai2's open-source olmOCR converts 1M PDF pages for under $200
thisguyknowsai · x · 2026-11-07
Ai2 open-sourced olmOCR, which converts PDFs, PNGs, and JPEGs into clean Markdown that preserves document structure — for under $200 per million pages.
- Built on a 7B vision-language model, it reads pages the way a person does instead of scraping the text layer
- Handles tables, equations, handwriting, multi-column layouts, figures and insets, and natural reading order
- Strips headers and footers so they don't pollute downstream chunks
- Runs locally on a single NVIDIA GPU with 12GB+ VRAM, or against a remote vLLM endpoint
- Ships with olmOCR-Bench, a benchmark suite of 7,000+ test cases across 1,400 documents
More from coding & agent
- LoRA Loom is a Windows desktop app for one-click local LoRA training — Abject-Ad1903 · 2026-11-07
- Which agent called which tool? Reddit thread on agent identity and verifiable logs — fred_pcp · 2026-11-07
- Computer-Using World Model Lets GUI Agents Simulate Actions Before Executing Them — burkov · 2026-11-07
- opentel-mcp v0.18 Adds SHA-256 Hashing to Detect MCP Tool 'Rug Pull' Attacks — Thirumalaiboobathi · 2026-11-07
- Uncle Bob on constraining coding agents with coverage, complexity and mutation testing — bibryam · 2026-11-07
- User asks his agent "can't or won't?" — and the agent backs down — StewartalsopIII · 2026-11-07