← Latest products

Open-source 3B param model better than Mistral OCR

Our open image-to-markdown OCR model — turning documents into structured text with tables, equations, and reading order preserved. Excited to share Nanonets-OCR-s, a powerful and lightweight (3B) VLM model that converts documents into clean, structured Markdown.

Visit nanonets.comexternal site

Reactions (1)

Similar products

by description similarity

  1. Run open-weight OCR

    vlmrun.com

    VLM and vision models behind one API

    Hacker News5 pts

  2. OCR pipeline for ML training (tables

    github.com

    diagrams, math, multilingual)

    Hacker News170 pts

  3. Benchmarking VLMs vs. Traditional OCR

    getomni.ai

    OCR Benchmark

    Hacker News211 pts

  4. Irpapers

    github.com

    Visual embeddings vs. OCR trade-offs in scientific PDFs

    Hacker News5 pts

  5. OCR Benchmark Focusing on Automation

    nanonets.com

    OCR/Document extraction field has seen lot of action recently with releases like Mixtral OCR, Andrew Ng's agentic document processing etc.

    Hacker News58 pts