Open Source
Explore the latest AI open-source projects from GitHub and HuggingFace.
Explore the latest AI open-source projects from GitHub and HuggingFace.
Explore the latest AI open-source projects from GitHub and HuggingFace. Discover projects across categories like LLM, Vision, Audio, and more.
492 projects
ruvnet
High-performance Rust vector and graph database for AI/agentic systems -- self-learning GNN layers, HNSW search, cognitive containers, and local LLM inference in one engine.
GVCLab
CVPR 2026 real-time portrait animation framework -- streamable diffusion for infinite-length live streaming with TensorRT acceleration and 12GB VRAM support.
Anthropic
Anthropic's official open-source Agent Skills framework for Claude -- 74.9k stars, standardized skill templates, cross-platform deployment across Claude Code, Claude.ai, and API.
sansan0
AI-driven trend monitoring tool with multi-platform aggregation, LLM-powered analysis, and 9+ notification channels -- 47.1k stars, v6.0.0 with MCP support.
AIDC-AI
AI-powered automated short video generation engine -- topic-to-video pipeline with scriptwriting, visual generation, TTS, and music composition. Multi-LLM support, ComfyUI-based.
Sesame AI Labs
Sesame AI Labs' open-source 1B-parameter conversational speech model using Llama architecture — natural human-like intonation, multi-speaker support, HuggingFace Transformers native.
Mindverse
Mindverse's open-source framework for training a personalized AI digital twin locally — full privacy, Hierarchical Memory Modeling, decentralized AI network support.
VectifyAI
Open-source vectorless RAG framework that uses LLM reasoning over hierarchical document trees instead of vector similarity search, achieving 98.7% accuracy on FinanceBench.
Muratcan Koylan
Comprehensive open-source collection of 13+ reusable AI agent skills for context engineering, cited in peer-reviewed research, working across Claude Code, Cursor, and other agent platforms.
NevaMind AI
Open-source memory framework for 24/7 proactive AI agents with hierarchical three-layer architecture, dual-agent design, and multi-provider LLM support.
CAMEL-AI
CAMEL-AI's multi-agent framework ranked #1 on GAIA benchmark (69.09%) with 30+ built-in toolkits including MCP, browser automation, and Gemini 2.5 Pro support.
OpenAI
Open-source LLM evaluation framework with 17.9k stars, providing benchmark registry, model-graded evals, and cross-model testing for systematic AI quality assurance.