Trending AI projects

CodeAF is an open‑source Go tool that turns LLMs into a multi‑project coding factory. From a single terminal window you can chat, have the system plan, run and check code across many repositories, using a crew of three model roles per task. It auto‑selects the cheapest open model, supports specialist sub‑harnesses, offers a headless CLI for CI, and runs with a tiny binary footprint.

A local-first, source-available AI meeting assistant that provides real-time transcription, context-aware answers, and advanced stealth features for interviews and professional calls.

Kimodo is a motion diffusion model that generates high-quality 3D animations for humans and robots, controlled via text prompts and precise kinematic constraints.

A command-line state machine that turns AI agent workflows into inspectable graphs of states and executable checks to ensure reliability in long-running tasks.

Occamy-1.0 is an open-source, compact agentic language model (35B total, 3B active) designed for long-horizon co-work tasks like tool use, file manipulation, and multi-step workflows. It achieves leading results on co-work benchmarks while keeping inference efficient.

A browser automation skill for AI agents that provides web connectivity, CDP-based browser control, and the ability to use existing browser sessions and local history.

Gravitas is an open-source reliability runtime for AI coding agents that enforces engineering discipline by requiring agents to ground their edits in evidence and verify their work through automated tests before finishing a task.

OpenEnv is an end-to-end framework for creating, deploying, and using isolated execution environments for agentic reinforcement learning, with a simple Gymnasium-style API. It provides a unified interface to diverse environments (games, coding sandboxes, browsers) as Docker images, enabling training with major RL frameworks.

A unified CLI and Python API that provides an easy-to-install wrapper for multiple Whisper AI transcription backends, featuring isolated environments and speaker diarization.

MAPIE is a Python library that provides model‑agnostic prediction intervals and risk‑controlled prediction sets for regression, classification, time‑series, and computer‑vision tasks. It works with any scikit‑learn estimator (and can wrap TensorFlow/PyTorch models) and offers statistically guaranteed uncertainty estimates based on conformal prediction. Install via `pip install mapie` and use the `MapieRegressor` or `MapieClassifier` wrappers to obtain calibrated intervals/sets with minimal code.

Prismor is a runtime security and governance platform for AI agents that monitors and enforces policies on agent actions, tool calls, and network access to prevent security risks like prompt injection and data exfiltration.

A curated collection of “Agent Skills” that teach how to manage an LLM’s limited context window. The repo organises the material into self‑contained skill folders (foundational, architectural, operational, etc.) that can be loaded by Claude Code, Cursor, or any Open‑Plugins host. It explains context‑engineering concepts (lost‑in‑the‑middle, compression, self‑editable context), provides platform‑agnostic examples, and includes benchmark results showing high skill‑router accuracy on frontier models.

An open-source wearable AI hardware development baseline providing stable APIs and reference implementations for building AI-powered wearable applications.

PaperQA2 is an agentic RAG system that enables high-accuracy, cited question answering over scientific papers, PDFs, and technical documents.

A 3D virtual office for AI agents that turns background automation into a visible, immersive workspace for real-time monitoring and collaboration.

A Go-based reverse proxy that automates the management and hot-swapping of local AI inference servers, allowing you to run multiple models seamlessly through a unified API.

Kauldron is a modular machine learning training library designed to increase research velocity by providing a flexible glue layer for datasets, models, and optimizers.

An open-source engine that allows AI agents to generate narrated, animated explainer videos locally by orchestrating text-to-speech, graphics rendering, and automated camera work.

XRAG is an open‑source benchmarking framework for Retrieval‑Augmented Generation (RAG). It lets users evaluate individual RAG components—retrievers, embeddings, LLMs, orchestrators—using classic IR metrics and LLM‑based quality scores. Features include a modular plug‑in architecture, support for OpenAI, Ollama, and Hugging Face models, a CLI, a web UI for interactive testing, and a REST API service. Install via pip, configure with a TOML file, and run benchmarks or generate synthetic QA data. Ideal for researchers and engineers building or comparing RAG systems.

A collection of SDKs and plugins that integrate Stripe's billing and payment infrastructure with LLMs and AI agent frameworks.

Hands-On Modern RL is an open-source book and course that guides learners from RL fundamentals like Markov decision processes and PPO to modern topics like RLHF, agentic AI, and multimodal RL, with runnable code and experiments throughout.

Sentient OS is an open‑source macOS personal AI that reads your local files, mail, calendar and messages, builds a Markdown knowledge base, and lets you hand off tasks via a shortcut. It can type, click, fill forms (Sidekick), draft replies in your voice (Double‑Tap), and surface proactive suggestions after overnight on‑device processing. You can run the built‑in model locally or plug in any compatible local or hosted LLM, and all raw data stays on your Mac unless you enable optional encrypted cloud sync.

HOL Guard is a Python‑based, local‑first antivirus for AI agents. It intercepts shell commands, file access, package installs, and plugin usage, flagging secrets, destructive actions, prompt‑injection, and supply‑chain risks. Users install via `pipx`, run a wizard to hook supported agents (Codex, Claude Code, Copilot CLI, Gemini, etc.), and manage approvals through a CLI or optional cloud dashboard. The repo also provides `plugin‑scanner` for CI security checks of AI plugins, with a ready GitHub Action.

An AI-assisted PCB design plugin for KiCAD 10 that allows AI assistants to directly create schematics, lay out PCBs, and run design audits using the Model Context Protocol.

An AI-native presentation app that generates editable slides from ideas, outlines, or documents using natural language editing and the nano banana pro model.

An open-source generative BI engine that provides a governed semantic layer for AI agents to turn business questions into correct SQL and deployable dashboards.

A tool for developers to track and optimize AI coding spend by analyzing local session files from over 40 AI coding agents.

A real-time open-vocabulary relation prediction system that identifies relationships between objects in images using custom-defined predicates without retraining.

NeuralNote is an audio plugin that uses the MuScriptor transformer model to transcribe audio files into editable MIDI data directly within your DAW, supporting multi-instrument detection and local-only processing.

InfraTech is a collection of educational resources and practical code exercises designed to help developers master AI infrastructure, including model training, inference optimization, and distributed systems.

A programmable TypeScript harness for building autonomous agents with built-in support for secure sandboxes, durable sessions, and tool integration.

A PyTorch-based toolkit for developing and deploying professional speech AI models, including ASR, TTS, and Speech LLMs.

Kimi K3 is an open-weight, 2.8T-parameter native multimodal MoE model designed for long-horizon coding, deep research, and complex agentic knowledge work.

A collection of AI agent skills that automatically generate interactive, single-file HTML animations for technical presentations, system diagrams, and product UI demos.

Midscene.js is a vision‑based GUI agent that lets developers write end‑to‑end tests for web, mobile, and desktop apps using natural‑language steps. It locates elements from screenshots, performs actions, and asserts visual outcomes across multiple platforms, supporting a variety of multimodal LLMs for low‑cost, human‑like testing.

HAPI is a local-first tool that wraps official AI coding agents (Claude Code, Codex, Cursor Agent, etc.) so you can run and control their sessions remotely from iOS/Android apps, a web/PWA client, or Telegram Mini App. It supports seamless local↔remote handoff, phone-based approvals, remote terminal access, voice control, and an optional workspace browser. Setup is two npx commands; the web client is free, native apps are paid, and the relay is end-to-end encrypted via WireGuard + TLS.

A local safety system for coding agents that prevents the execution of dangerous commands and blocks secret leaks through pre-execution blocking.

NVIDIA‑NeMo / Nemotron provides open‑source multimodal foundation models (Nano, Lightning, Super, Ultra, Omni) plus a modular “step” framework, full training recipes, deployment cookbooks, and examples for building agentic AI systems. All models and pipelines are released under Apache 2.0 and can run from edge GPUs to datacenter clusters.

EdenLive is an open‑source system that turns low‑fidelity reference videos (e.g., a game’s proxy rendering) into high‑quality, real‑time generative video. It provides 5 B and 14 B diffusion models, supports both bidirectional and chunk‑autoregressive inference, and includes a streaming WebRTC service, a text‑to‑game skill, and integration with Reactor and LightX2V for accelerated deployment.

A multimodal lakehouse and vector database built on the Lance columnar format for fast, scalable vector search across petabytes of multimodal data.

A pre-trained, enterprise-grade Voice Activity Detector (VAD) that identifies speech in audio files with high accuracy and low latency, suitable for edge and mobile deployment.

Heeler is a native iOS dashboard for the herdr agent-first terminal runtime, allowing developers to monitor, steer, and interact with coding agents on remote machines via SSH.

Thinking-orbs is a library of lightweight, 2D canvas-based loading animations designed to provide visual feedback for AI agents in web interfaces.

A distilled cybersecurity knowledge base of 102 skills across 14 domains, structured as a hierarchical routing system for AI agents to perform penetration testing and security research.

A large collection of Jupyter notebooks that teach how to build generative‑AI agents—from simple chatbots to multi‑agent business, creative, and research systems—using LangChain, LangGraph, AutoGen and related tools. Each notebook includes explanations, runnable code, and links to videos. The repo is open‑source, community‑driven, and serves as a step‑by‑step learning path for building production‑ready AI agents.

smolvm is an open‑source micro‑VM runtime that lets AI agents run in fast, isolated VMs. It supports branchable checkpoints, portable `.smolmachine` bundles, and SDKs for Python, Node, and Rust, enabling safe sandboxing and reproducible state for AI workloads.

A TypeScript framework for building autonomous agents that share a unified action layer, database, and state with a purpose-built user interface.

An open-source framework for building general AI agents capable of browser automation and data analysis.