Open Source AI Projects
Discover 87+ curated GitHub repositories for building with AI. Updated daily with trending projects, frameworks, and tools.
Showing 21–40 of 87 projects
Page 2 of 5
lobe-chat
lobehub/lobe-chat
A modern AI agent workspace that supports multiple AI providers, RAG-based knowledge bases, and one-click deployment with MCP marketplace integration. Built with TypeScript/NextJS, it provides a comprehensive platform for deploying private AI agent applications.
Offers production-ready AI agent deployment with excellent multi-provider support and modern UI, perfect for teams wanting to quickly deploy custom AI assistants.
Prompt-Engineering-Guide
dair-ai/Prompt-Engineering-Guide
Comprehensive educational resource covering prompt engineering, context engineering, RAG, and AI agents through guides, papers, lessons, and notebooks. Written in MDX format for easy navigation and learning.
Essential reference for mastering prompt engineering and AI agent techniques with curated, up-to-date content from industry experts.
OpenHands
OpenHands/OpenHands
AI-driven development platform that automates coding tasks using various LLMs including ChatGPT and Claude. Provides CLI and developer tools for autonomous software development assistance.
Transforms development workflow by automating repetitive coding tasks and providing intelligent code assistance across multiple AI models.
LlamaFactory
hiyouga/LlamaFactory
Unified fine-tuning framework supporting 100+ LLMs and VLMs with advanced techniques like LoRA, QLoRA, and RLHF. Provides efficient training tools for instruction tuning and model customization.
Streamlines the complex process of fine-tuning large language models with state-of-the-art techniques and broad model support.
scikit-learn
scikit-learn/scikit-learn
Mature Python machine learning library providing simple and efficient tools for data mining, data analysis, and predictive modeling. Built on NumPy, SciPy, and matplotlib with comprehensive algorithms.
Industry standard for traditional machine learning with excellent documentation, stability, and wide adoption across data science teams.
MetaGPT
FoundationAgents/MetaGPT
Multi-agent framework that simulates an AI software company, enabling natural language programming through coordinated AI agents. Focuses on collaborative AI development workflows.
Pioneering approach to software development automation using coordinated AI agents, ideal for exploring next-generation development paradigms.
OpenBB
OpenBB-finance/OpenBB
Financial data platform providing comprehensive APIs and tools for analysts, quants, and AI agents to access market data, economics, and quantitative finance information. Built in Python with extensive financial datasets.
Essential toolkit for financial AI applications with robust data access and analysis capabilities specifically designed for quantitative finance.
pathway
pathwaycom/pathway
Python ETL framework optimized for stream processing, real-time analytics, and LLM pipelines with built-in RAG capabilities. Combines batch and streaming data processing with AI-first design.
Bridges traditional data processing with modern AI workflows, perfect for building real-time RAG systems and LLM data pipelines.
autogen
microsoft/autogen
Microsoft's programming framework for building agentic AI systems with multi-agent collaboration capabilities. Provides structured approaches to agent communication and task orchestration.
Industry-leading framework from Microsoft for enterprise-grade agent development with strong community support and documentation.
anything-llm
Mintplex-Labs/anything-llm
All-in-one desktop and Docker AI application with built-in RAG, AI agents, no-code agent builder, and MCP compatibility. Provides complete AI workspace with local deployment options.
Perfect for teams wanting a complete AI solution with both technical flexibility and no-code options for rapid agent development.
ai-agents-for-beginners
microsoft/ai-agents-for-beginners
Educational curriculum with 12 structured lessons for learning AI agent development using Microsoft's frameworks including AutoGen and Semantic Kernel. Provides hands-on Jupyter notebook exercises.
Best starting point for developers new to AI agents with structured learning path and practical exercises from Microsoft.
Flowise
FlowiseAI/Flowise
Visual no-code platform for building AI agents and workflows using drag-and-drop interface. Built with TypeScript and integrates with LangChain for comprehensive agent development.
Democratizes AI agent creation for non-technical users while maintaining flexibility for developers through visual workflow design.
llama_index
run-llama/llama_index
Leading framework for building LLM-powered agents over custom data with comprehensive RAG capabilities, vector database integration, and multi-agent support. Provides production-ready tools for data-driven AI applications.
Industry standard for RAG applications with excellent documentation, active development, and comprehensive tooling for data-centric AI.
whisper.cpp
ggml-org/whisper.cpp
High-performance C++ implementation of OpenAI's Whisper speech recognition model optimized for inference. Provides efficient speech-to-text capabilities without Python dependencies.
Essential for production speech recognition with minimal resource requirements and cross-platform compatibility.
mem0
mem0ai/mem0
Universal memory layer for AI agents providing persistent memory management, long-term context retention, and state management. Enables AI applications to maintain context across sessions.
Solves critical memory limitations in AI agents by providing persistent, searchable memory for enhanced conversational AI experiences.
JeecgBoot
jeecgboot/JeecgBoot
AI-powered low-code platform for enterprise applications with integrated AI features, code generation, and workflow automation. Built with Java/Spring Boot and supports AI model integration.
Accelerates enterprise development with AI-powered code generation and comprehensive low-code capabilities for business applications.
crewAI
crewAIInc/crewAI
Framework for orchestrating role-playing autonomous AI agents that collaborate on complex tasks. Focuses on agent specialization and team-based problem solving through structured workflows.
Simplifies multi-agent system development with role-based collaboration, perfect for complex business process automation.
milvus
milvus-io/milvus
High-performance, cloud-native vector database built for scalable similarity search and AI applications. Provides distributed architecture with support for billion-scale vector operations.
Production-grade vector database essential for large-scale RAG and similarity search applications with enterprise reliability.
gradio
gradio-app/gradio
Python framework for building and sharing machine learning web interfaces with minimal code. Provides interactive UI components for ML model deployment and data visualization.
Fastest way to create web interfaces for ML models and data science projects with zero frontend development required.
siyuan
siyuan-note/siyuan
Privacy-first, self-hosted personal knowledge management system with AI integration including ChatGPT, DeepSeek, and Ollama support. Built with TypeScript and Golang for local-first operation.
Ideal for developers wanting AI-enhanced note-taking with complete data privacy and self-hosting capabilities.