Hub / Инструменты
Инструменты
60 записей, по убыванию рейтинга. Как считается рейтинг.
- firecrawl/firecrawlSupercharge your AI agents with data from the web and beyond. Building the library for superintelligence. 🔥рейтинг 79.9 · 199 850 звёзд · AGPL-3.0 · сообщество
- landing-ai/ade-cliThe official CLI for Agentic Document Extraction (ADE) by LandingAI — parse documents and extract schema-shaped data from your terminalрейтинг 78.7 · 2 424 звёзд · Apache-2.0 · сообщество
- datalab-to/indra1M param OCR model - runs on CPU, 15x faster than tesseract, close to 1B+ param models in accuracyрейтинг 77.8 · 1 687 звёзд · Apache-2.0 · сообщество
- go-rod/rodA Chrome DevTools Protocol driver for web automation and scraping.рейтинг 76.1 · 7 146 звёзд · MIT · сообщество
- Topdu/OpenOCROpenOCR: An Open-Source Toolkit for General-OCR Research and Applications, integrates a unified training and evaluation benchmark, commercial-grade OCR and Document Parsing systems, and faithful reproductions of the core implementations from a wide range of academic papers.рейтинг 75.9 · 1 495 звёзд · Apache-2.0 · сообщество
- FengZhenfei/carrelOntology-Augmented Generation for AI agents. рейтинг 75.9 · 801 звёзд · MIT · сообщество
- aiptimizer/TurboOCRTurboOCR, >200 img/s OmnidocBench. TensorRT FP16, PP-OCRv6, HTTP + gRPCрейтинг 75.8 · 1 158 звёзд · MIT · сообщество
- enoch3712/ExtractThinkerExtractThinker is a Document Intelligence library for LLMs, offering ORM-style interaction for flexible and powerful document workflows.рейтинг 75.6 · 1 604 звёзд · Apache-2.0 · сообщество
- ggozad/haiku.ragAgentic RAG for local and self-hosted document search: hybrid retrieval, reranking and multimodal RAG on embedded LanceDB, with Docling parsing and an MCP serverрейтинг 75.4 · 654 звёзд · MIT · сообщество
- alirezamika/autoscraperA Smart, Automatic, Fast and Lightweight Web Scraper for Pythonрейтинг 75.3 · 8 078 звёзд · MIT · сообщество
- run-llama/ParseBenchParseBench - A Document Parsing Benchmark for AI Agentsрейтинг 75.3 · 626 звёзд · Apache-2.0 · сообщество
- skygazer42/MimirQ中文优先的企业 RAG 知识库：可控解析、治理、切块、混合检索、重排、引用、图谱、评测与 Dify 接入。рейтинг 75.2 · 593 звёзд · Apache-2.0 · сообщество
- neosun100/DeepSeek-OCR-WebUISelf-hosted DeepSeek-OCR-2 WebUI & API — vLLM / transformers on NVIDIA, Apple Silicon & CPU, per-page quality checks, PDF streaming, MCPрейтинг 74.5 · 446 звёзд · MIT · сообщество
- getmaxun/maxunTurn any website into a structured API. Extract, automate, search and monitor the web.рейтинг 73.8 · 18 177 звёзд · AGPL-3.0 · сообщество
- scub-france/docling-StudioVisual document analysis studio powered by Docling — configure the extraction pipeline, inspect text, tables and bounding boxes in the browser, then chunk, embed and index into OpenSearch and Neo4j.рейтинг 73.2 · 271 звёзд · MIT · сообщество
- arthurpanhku/DocSentinelMCP server for AI agent for cybersecurity: automate assessment of documents, questionnaires & reports. Multi-format parsing, RAG knowledge base,Risks, compliance gaps, remediations. рейтинг 70.4 · 90 звёзд · MIT · сообщество
- NameetP/pdfmuxPDF extraction that audits its own output — and certifies any other extractor's, catching pages they silently dropped. Verify signed manifests offline: free, MIT, no account. 0.903 on opendataloader-bench, #2 of 8 engines. 7-tool MCP server.рейтинг 70.3 · 86 звёзд · MIT · сообщество
- unitedrhino/doclingGet your documents ready for gen AI, in pure Go. A Go alternative to anydoc & Docling — 纯 Go 实现的文档解析库:PDF/Word/PPT/Excel/HTML/Markdown/EML/图片等 12 种格式统一解析为 Docling 协议(Markdown/JSON/content_list),表格还原·多栏阅读顺序·图表数据回填,OCR 收敛为一个大模型接口,单二进制零依赖,快 34~153 倍。联犀物联网平台知识库同款引擎。рейтинг 70.1 · 78 звёзд · MIT · сообщество
- lorien/awesome-web-scrapingList of libraries, tools and APIs for web scraping and data processing.рейтинг 67.8 · 8 221 звёзд · NOASSERTION · сообщество
- YOLOW92/pagesiftSelf-hosted web scraper, crawler and document parser for AI agents: URLs, PDFs with OCR and Office files to clean Markdown/JSON. Firecrawl- and Jina-compatible API, MCP server, agent endpoint, Python/Node SDKs. Single Rust binary.рейтинг 63 · 255 звёзд · AGPL-3.0 · сообщество
- cporter202/API-mega-listThis GitHub repo is a powerhouse collection of APIs you can start using immediately to build everything from simple automations to full-scale applications. One of the most valuable API lists on GitHub—period. 💪рейтинг 61.9 · 8 326 звёзд · лицензия не указана · сообщество
- ComPDFKit/docslightWith DocSlight, precisely parse and extract data from any document, including PDFs, scans, images, and Office files. It is an open-source AI project from ComPDF (KDAN ecosystem).рейтинг 61.5 · 138 звёзд · LGPL-3.0 · сообщество
- flamehaven01/Flamehaven-FilesearchSelf-hosted RAG search engine — 34 formats, BM25+hybrid search, multi-LLM (Gemini/OpenAI/Claude/Ollama), FastAPI + Docker, production-ready in 3 minрейтинг 58.1 · 109 звёзд · MIT · сообщество
- stanford-oval/ChurroCHURRO is an OCR toolkit for historical document transcription, built to make handwritten and printed sources readable at high accuracy and lower cost.рейтинг 54.7 · 86 звёзд · Apache-2.0 · сообщество
- NanoNets/docstrangeExtract and convert data from any document, images, pdfs, word doc, ppt or URL into multiple formats (Markdown, JSON, CSV, HTML) with intelligent structured data extraction and advanced OCR.рейтинг 47.7 · 1 586 звёзд · MIT · сообщество
- AdemBoukhris457/Documents-Parsing-LabJupyter notebooks testing different OCR models for document parsing (Dolphin, MonkeyOCR, Marker, Nanonets, ...)рейтинг 44 · 82 звёзд · лицензия не указана · сообщество
- LeDat98/NexusRAGHybrid RAG system combining vector search, knowledge graph (LightRAG), and cross-encoder reranking — with Docling document parsing, visual intelligence (image/table captioning), agentic streaming chat, and inline citations. Powered by Gemini or local Ollama models.рейтинг 39.4 · 530 звёзд · лицензия не указана · сообщество
- InternLM/Visual-ERM[NeurIPS 2026] Official Implementation of "Visual-ERM: Reward Modeling for Visual Equivalence"рейтинг 31.7 · 68 звёзд · лицензия не указана · сообщество
- harishdeivanayagam/rowfillOpen-source spreadsheets platform for deep research and document processingрейтинг 29.9 · 364 звёзд · NOASSERTION · сообщество
- CycloneBoy/pdf_tableA Unified Toolkit for Deep Learning-Based Table Extractionрейтинг 19.4 · 60 звёзд · лицензия не указана · сообщество