Unstract
No-code platform turning PDFs, scans and images into structured JSON with LLMs. Define the schema in plain English, deploy as a REST API or ETL pipeline.
// LLM
10 entries tagged with "llm"
// ENTRIES
10 entries
Hindsight
Agent memory system that learns instead of just storing. Fact extraction, multi-strategy recall and reflection, self-hosted in Python with client SDKs.
Markdownify MCP
MCP server converting PDFs, images, audio, DOCX, XLSX, PPTX, web pages and YouTube videos into clean Markdown, ready to feed straight into an LLM context.
PaddleOCR
Open-source OCR toolkit that turns images and PDFs into structured, LLM-ready data (Markdown, JSON). Handles 100+ languages, tables, formulas and charts.
Receipt OCR
Python OCR engine for receipts: raw text extraction via Tesseract and structured data via LLMs (OpenAI, Gemini, Groq). Ships with a CLI and FastAPI.
Flowise
Visual no-code builder for AI agents and workflows with drag-and-drop. LangChain connectors, RAG, multi-agent, LLM integrations, REST API and Docker self-host.
MarkItDown
Microsoft's Python library and CLI to convert PDF, Word, Excel, images, audio and more into clean Markdown, ready for LLMs and text analysis pipelines.
Agent Reach
A CLI scaffolding that gives AI agents eyes on the whole internet: read and search Twitter, Reddit, YouTube, GitHub, Bilibili, RSS with zero paid APIs.
Dify
Open-source platform to build LLM apps: visual workflows, RAG pipeline, agents, model management and observability, from prototype to production.
Hermes Agent
Self-improving AI agent from Nous Research: terminal UI, messaging bridges, cron scheduler, autonomous skill creation and multi-LLM support, VPS to serverless.