Open Source AI Projects
Artificial intelligence (AI) refers to systems that perform tasks such as recognizing patterns, generating content, making predictions, and interpreting language. AI tools can automate repetitive work, help people analyze large amounts of data, and add intelligent features to applications. They range from models that run locally to services and libraries that connect AI capabilities with software, devices, and creative workflows.
Open source AI resources include machine-learning frameworks, model runtimes, agent systems, coding assistants, and tools for image, audio, and text processing. When choosing one, consider its license, maintenance activity, documentation, hardware and software requirements, data handling, and compatibility with your existing stack. Some tools are intended for experimentation, while others support production use. They can be useful to developers, researchers, organizations, and individuals building or adapting AI systems.
382 repositories · updated October 3, 2026
colibri: Run Large MoE Models on Local Hardware
colibri is a C inference engine that streams Mixture-of-Experts model weights from disk, so large models can run without fitting entirely in RAM or VRAM. It supports CPU-only use and optional GPU backends, with speed depending on hardware and storage.
Open-Higgsfield-AI: Review and Run an AI Image Studio
A JavaScript-based AI image studio presented as a self-hosted alternative to Higgsfield AI. The repository currently describes the app as an internal review tool, with image generation paused until provider credits are available.
open-pencil: Edit and Automate Design Files
OpenPencil is an AI-assisted design editor for opening and editing .fig and .pen files. It combines a visual canvas with CLI, MCP, and scripting tools for design workflows, plus real-time collaboration and a headless Vue SDK.
web-design: A Claude Code SKILL for Spec-First Web Page Design
The web-design project is a Claude Code SKILL designed to streamline the creation of beautiful and consistent web pages. It emphasizes a 'spec first, code second' approach, ensuring design principles are established before development begins. This tool helps generate UI, visuals, motion, and responsiveness that are consistent across pages and easily editable.
OrcaReplay: Time Travel for AI Agents, Debugging and Evaluation
OrcaReplay introduces "time travel" capabilities for AI agents, allowing developers to record, replay, fork, and debug any agent run with any model. It addresses the challenges of AI agent debugging by providing byte-for-byte reproducibility, offline analysis, and the ability to compare different models from specific checkpoints. This tool, built by the OrcaRouter.ai team, enhances observability and control over complex agent behaviors.
Shepherd: Reversible Execution Traces for Programmable Meta-Agents
Shepherd is a Python runtime substrate designed for agent work requiring inspection, reversibility, and supervision. It records agent runs as durable, inspectable execution traces, enabling meta-agents to observe, fork, replay, and revert any operation. This framework couples agents and environments using a copy-on-write fork, offering significant performance benefits and robust permission enforcement.
SwarmLLM: Run Local AI Models and Team Up for Giant Distributed Inference
SwarmLLM is a free, open-source application that allows you to run AI chat models directly on your own computer. It uniquely enables multiple computers to team up over the internet, collectively running models too large for a single machine. This platform offers an OpenAI and Anthropic-compatible API, all without requiring accounts or cryptocurrency.
Agent Anvil: CI-First Evaluation Harness for Tool-Using AI Agents
Agent Anvil is a robust, CI-first evaluation harness designed for AI agents that utilize tools. It meticulously runs scenario suites, captures detailed traces of agent behavior, and provides semantic grading to identify issues. The platform excels at clustering failures and suggesting concrete fixes for prompts, tools, and guardrails, ensuring agents behave safely and effectively.
CCCC: Coordinate Your Coding Agents Like a Group Chat
CCCC is a production-minded orchestrator designed to coordinate coding agents like a group chat, offering features such as read receipts, delivery tracking, and remote operations from your phone. It enables 24/7 workflow for multi-agent teams with a single `pip install` and zero infrastructure. This tool helps manage diverse AI runtimes, ensuring persistent collaboration across different machines and trusted working groups.
AI Session Search: Ultra-Fast AI Agent Session Analysis
AI Session Search (aise) is an ultra-fast, Rust-powered tool designed for searching and analyzing local AI agent coding sessions. It seamlessly integrates and indexes nine different session formats, including those from Claude, Codex, Cursor, and Gemini CLI. This powerful utility enables developers to quickly recover context, track agent behavior, and efficiently manage their AI-generated code history.
Pydantic AI Harness: Enhancing Your AI Agents with Robust Capabilities
Pydantic AI Harness is the official capability and harness library for Pydantic AI, designed to extend agents for complex, long-running tasks. It provides a modular system of "capabilities" for functionalities like file system interaction, web research, memory, and sub-agent delegation. This library enables developers to build sophisticated and durable AI agents with ease.
Router: Optimize AI Model Selection and Costs for Agentic Systems
The Weave-OS Router is an intelligent model router for agentic systems, optimizing AI model selection for every request. It acts as a drop-in proxy for major AI providers, routing prompts to the most suitable model in under 50ms. This solution helps users significantly cut costs, often by 40-70%, simply by changing an endpoint.