Mathpix

Tiny open vision-language model for efficient image understanding.

An open-weights family of small vision-language models for captioning, visual Q&A, pointing, counting, and object detection — small enough to run on-device (checkpoints down to 0.5B on Hugging Face). Run it locally with the Photon engine, or call Moondream Cloud's OpenAI-compatible API with a free monthly credit tier and pay-per-image pricing.

Agentic document parsing and extraction for AI teams, via one API.

A document-intelligence API that parses, splits, extracts, and edits PDFs, images, spreadsheets, and slides into clean, structured output for RAG and AI pipelines. It blends custom in-house models with frontier ones and bills via usage credits, automatically discounting pages it can parse without the heavier pipeline.

AI agents for document processing and enterprise data extraction.

Nanonets automates document-heavy workflows — invoices, orders, contracts, and claims — with AI agents that read, extract, and route structured data across ERPs, email, and approval chains. It runs on its own OCR-3 extraction model and can fold in LLMs for agentic pipelines. Offered as managed cloud with VPC, single-tenant, and on-premises deployment options and regional data residency.

Toolkit that turns documents into AI-ready Markdown and JSON.

A document-processing toolkit that converts PDF, DOCX, PPTX, XLSX, HTML, images, and audio into clean Markdown or JSON for LLM and RAG pipelines. It does advanced PDF understanding — page layout, reading order, table structure, and OCR for scans — and ships a hybrid chunker plus native LangChain and LlamaIndex integrations. Small enough to run on a laptop via a Python API or CLI; MIT-licensed and community-governed.