HomeProjectsArticlesPK BoardAbout
Explore Projects

Sparrow

Active
GitHub Python GPL-3.0

Description

Sparrow is a structured data extraction tool that supports instruction calling with ML, LLM, and Vision LLM for extracting structured information from documents, suitable for document parsing in RAG pipelines.

Key Features

  • Extract structured data from invoices, receipts, statements, forms and tables
  • Pluggable architecture mixing Vision LLM, Text LLM, and Agent pipelines
  • Multiple backends: MLX (Apple Silicon), vLLM (NVIDIA), Ollama, Hugging Face
  • JSON Schema-based automatic validation ensures extraction results meet expected structure
  • RESTful API-first design for easy integration into any backend or data pipeline
  • Built-in multi-step agent workflow orchestration and visual monitoring dashboard

Use Cases

💡 Automated structured data extraction from enterprise invoices and financial statements
💡 Convert scanned documents and PDFs to searchable structured JSON for RAG pipelines
💡 Batch processing and validation of bank statements and tabular data
💡 Orchestrate complex multi-step document processing workflows via the agent framework

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (5.2k stars)
  • • Established track record (4 years in production)
  • • Responsive to issues, low backlog

⚠️ Limitations

  • • Restrictive license (GPL-3.0)

Tags

data-extraction document-processing llm vision rag python

Categories

📚 RAG Tools

Quick Start

Install Python 3.12.10+, clone the repo, and cd into sparrow-ml/llm. Install requirements_sparrow_parse.txt. macOS users need brew install poppler first. Start the API server, then use sparrow.sh with a JSON Schema and document path to extract structured data.
Visit GitHub Visit Website

Related Projects

MemAgent

1.1k · Python
Stale B

A MemAgent framework that can extrapolate to 3.5M context tokens, along with a training framework for RL training of any agent workflow.

memoryagentrag +2
  • · Ultra-Long Context Processing: Extrapolate from 8K training context to 3.5M tokens with performance loss under 5%
  • · Reinforcement Learning Driven: Trained with RLVR (Reinforcement Learning from Verifiable Rewards), extends DAPO algorithm for end-to-end multi-turn conversation optimization
  • · Linear Time Complexity: Breaks through computational bottlenecks in long-text processing with linear resource scaling
View on GitHub

LightRAG

38.9k · Python
Active A+

LightRAG is a simple and fast Retrieval-Augmented Generation framework using graph-enhanced retrieval, published at EMNLP 2025.

raggraphretrieval +2
  • · Graph-enhanced retrieval — Dual-level retrieval (local/global) using knowledge graph entities and relationships, more precise than vector search
  • · Four text chunking strategies — Fixed, Recursive, Vector, and Paragraph chunking adapted to different document types
  • · Multi-backend storage — Neo4j, PostgreSQL, MongoDB, OpenSearch, JSON KV Store and other storage backends supported
View full details → View on GitHub

Quivr

39.4k · Python
Stale C

Opinionated RAG framework for integrating GenAI into your apps. Works with any LLM, any vectorstore, any files — so you can focus on your product instead of building RAG pipelines.

ragpythonvector-database +3
  • · Opinionated RAG framework — Ready-to-use RAG pipeline, no need to build from scratch
  • · Any file support — PDF, TXT, Markdown and more with custom parser support
  • · Multi-LLM compatible — OpenAI, Anthropic, Mistral, Gemma and Ollama local models
View full details → View on GitHub

Awesome LLM Apps

132.9k · Python
Active A+

100+ AI Agent and RAG apps you can actually run — clone, customize, and ship. A great reference for quickly building LLM-powered applications.

agentragllm +1
  • · 100+ runnable templates — Original, end-to-end tested projects, not a curated link collection
  • · 14 categories — AI Agents, Multi-agent, MCP Agents, Voice AI, RAG, Agent Skills, Fine-tuning, and more
  • · Multi-model compatible — Supports Claude, Gemini, OpenAI, xAI, Qwen, Llama with config-only switching
View full details → View on GitHub