Trending AI projects

Python bindings for llama.cpp that enable local LLM inference with an OpenAI-compatible API and support for multiple hardware acceleration backends.

An extension for the Pi terminal AI editor that adds a customizable status bar, persistent bash mode, prompt queue management, and enhanced workflow shortcuts.

A self-hosted, open-source photo management platform that uses local machine learning models for face recognition, object detection, and semantic search.

An AI-powered recursive learning system that turns AI coding assistants into interactive tutors using structured knowledge maps and TDD-style exercises.

A multi-device, persistent terminal built for coding agents that allows users to run AI coding sessions across desktop and mobile with server-side session recovery.

BiliSum is a video-to-knowledge tool that converts Bilibili, YouTube, and local videos into transcriptions, visual notes, and a local RAG-powered knowledge base.

A reference RAG chat application that allows users to interact with their own documents using Azure OpenAI and Azure AI Search.

haiku.rag is a Python library/CLI for building a private, citation‑rich RAG system over PDFs, Office files, HTML, Markdown and images. It uses LanceDB for hybrid vector/full‑text indexing, Docling for document parsing, and Pydantic AI for LLM/embedding providers (defaulting to Ollama). Features include hybrid search, optional reranking, multimodal image search, sandboxed Python for data aggregation, and a CLI/TUI that shows exact page‑level citations. It offers an optional background ingester service and an MCP server for integration with tools like Claude Code. Install via `pip install haiku.rag` (or the slim core) and interact via commands like `haiku-rag add-src`, `haiku-rag ask`, or the async Python API.

An open-source platform for building and deploying AI assistants that answer from organizational knowledge with built-in routing, human escalation, and analytics.

A Python library for building predictable AI workflows by turning typed functions into structured LLM calls with automatic validation and retries.

LimiX is a foundation model for structured tabular data that performs classification, regression, and missing-value imputation without task-specific training.

A modular, production-grade SDK for connecting AI agents to WeChat via the iLink protocol, supporting Node.js, Python, Go, and Rust.

A GPU-accelerated library for vector search and clustering, providing high-performance approximate nearest neighbors implementations for AI and data mining.

A world model that generates real-time, photorealistic multi-camera video for autonomous driving simulation using HD maps and trajectory poses.

OpenCursor is an open-source AI coding agent for VS Code that allows developers to integrate various cloud or local models to automate coding tasks, file edits, and terminal operations with full review control.

vla‑evaluation‑harness is a Python library that lets you evaluate Vision‑Language‑Action models on dozens of robot‑simulation benchmarks (LIBERO, CALVIN, SimplerEnv, etc.) using Docker‑isolated environments and lightweight model‑server scripts. It provides batch‑parallel execution (sharding episodes + GPU batching) for huge speed‑ups, a unified leaderboard of 2 456 models, and a simple CLI/Python API for both research and training‑time evaluation.

A comprehensive, automated toolkit for AI video prompt engineering that provides verified prompt formulas, model-specific methodologies, and Claude-integrated skills for 24 different video generation models.

acestep.cpp is a C++‑based local server that generates 48 kHz stereo music from text prompts. It uses GGML‑quantized models (LM, text encoder, diffusion DiT, VAE) and runs on CPU, CUDA, Vulkan or Metal. After downloading the four GGUF model files, you build the binary (CPU‑only, CUDA, Vulkan, or all‑in‑one) and start the server. A web UI at http://localhost:8085 lets you write a caption, select a model/LoRA adapter, and generate, play, or download tracks. The server also offers a tiny REST API (POST /lm, /synth, /understand, /vae) and CLI tools for scripting. It is a genuine open‑source AI music‑generation project, not a tutorial list.

Google’s ADK‑Java is a Maven‑distributed Java library for building, testing, and deploying LLM‑powered AI agents. It offers a code‑first builder API, a rich tool ecosystem (including Google Search), multi‑agent composition, a web‑based development UI, and integration with Google’s A2A protocol for remote agent communication.

An open-source long-term memory layer for AI agents that enables persistent storage and recall of user preferences and conversational context across sessions.

A native-Rust AI agent workspace that allows users to run multiple isolated agents for coding and automation locally on their own machine.

A collection of Docker Compose templates that simplify the deployment and testing of various AI agent frameworks by automating environment and model configuration.

An MCP server that provides AI agents with real-time, accurate data on NixOS packages, options, and documentation to prevent hallucinations.

PyGPT is a versatile, open-source desktop AI assistant that provides a unified interface for chatting with various LLMs, executing multi-agent workflows, performing RAG on local files, and running system-level tools.

An AI-powered design skill that transforms architectural and cultural photos into art-directed editorial posters with integrated Chinese typography and spatial awareness.

RAGEN is a reinforcement learning framework for training reasoning LLM agents over multiple interaction turns, built on the StarPO algorithm. It includes 10 built-in environments, supports PPO and GRPO optimization, and provides diagnostics for failure modes like the Echo Trap and reasoning collapse. RAGEN-2 adds SNR-Adaptive Filtering and mutual-information-based metrics to detect template collapse.

An open-source driver assistance system based on a fork of openpilot that provides modified driving assist behaviors for over 300 car models.

pi Desktop is a cross‑platform GUI for the pi AI coding agent, letting you run, review, and manage coding sessions with timeline visualisation, Git diff review, built‑in terminal, parallel workers, theming, and an Android companion app, all sharing the same `~/.pi/agent` state as the CLI.

Late is an AI coding agent that prevents context degradation by delegating tasks to ephemeral subagents, ensuring the primary orchestrator remains focused and efficient during long, complex workflows.

Swiftide is a Rust framework for building LLM applications, offering tools for agent orchestration, typed task graphs, and streaming RAG pipelines.

OllamaFreeAPI is a Python client that gives free, no‑signup access to several open‑source LLMs (LLaMA, Mistral, DeepSeek, Gemma‑based models, etc.) via simple `chat` and `stream_chat` calls. It abstracts a set of community‑run API endpoints, offering model listing, info lookup, and basic load‑balancing. Ideal for quick experiments or learning, but relies on the reliability of volunteer‑hosted servers.

A molecular design tool that uses reinforcement learning and transfer learning to generate optimized small molecules based on user-defined property profiles.

Horosa Skill is a local-first MCP server and CLI that provides AI agents with offline, verifiable access to 110 professional astrology and metaphysics calculation techniques.

The XeSS 3 SDK is a set of AI-based technologies for real-time rendering that boosts frame rates and reduces input lag through super resolution and frame generation.

Jido is an Elixir‑native autonomous‑agent framework. It defines immutable agents with a `cmd/2` contract, actions that transform state, and directives that describe runtime‑owned side‑effects. Built on GenServer and OTP supervision, it supports multi‑agent hierarchies, signal routing, plugins, and execution strategies (direct or FSM). AI/LLM integration is optional via companion packages (`jido_ai`, `req_llm`). Install via Hex, define agents/actions, and run them under `AgentServer` for production‑grade reliability.

YOLO is an Obsidian community plugin that embeds AI agents directly into the note‑taking app. It provides chat, multi‑model support, retrieval‑augmented generation from your vault, tool‑calling, a learning mode with flashcards, multi‑window chats, memory, and a fast AI‑whiteboard. Install via the plugin store, configure an API key (OpenAI, Claude, Gemini, etc.), and start using Quick Ask (`@`) or the sidebar. The project is MIT‑licensed and actively maintained.

langchain-rust is a Rust implementation of LangChain, providing a framework for building LLM-powered applications with composable chains, agents, memory, document loaders, and integrations with many LLMs, embedding models, and vector stores.

A Rust-implemented, sandboxed bash interpreter with a virtual filesystem, designed to be a safe, introspectable tool for AI agents.

Dryforge is a plugin for AI‑code‑generation agents (Claude Code, Codex, Grok Build, Copilot, Antigravity) that flips the usual workflow: you first state your intent, the tool records it in plain Markdown, then the agent works autonomously inside that boundary. It provides three commands – `ready` (capture intent), `go` (realize and verify), and `migration` (bootstrap an existing repo). All decisions stay documented, the repo stays Git‑controlled, and the plugin works across multiple agent platforms.

Termany is a specialized terminal emulator designed to manage multiple concurrent AI coding agent sessions, providing a unified interface for monitoring, reviewing agent output, and handling project workflows.

A multi-agent framework that improves agent collaboration by treating the system as a unified recursive computation in latent space, reducing token usage and speeding up inference.

Agent Storyboard is a local‑first Node.js web app that lets Codex or Claude Code agents generate full video storyboards—script, shot list, AI‑generated images, motion clips, and voice‑overs—directly into a single spreadsheet‑style table. All data and media stay on your computer; the agent interacts via MCP calls, not a browser. Features include drag‑and‑drop shot ordering, batch generation, built‑in visual styles, Whisper‑based voice alignment, and export to Markdown/HTML/Word.

An AI-powered 3D model studio that converts images into interactive 3D models using multiple AI providers and provides a professional inspection and presentation workspace.

AI Gist is a privacy-focused, cross-platform prompt management tool that allows users to organize, template, and refine AI prompts while keeping all data stored locally.

A formal language for writing and verifying mathematical proofs based on set theory, designed to be more intuitive and intuitive than traditional formal verification tools.

pg_durable is a PostgreSQL extension that enables fault-tolerant, long-running background workflows directly inside the database, eliminating the need for external job queues or orchestrators.

A workflow orchestration engine for Pi that enables the execution of repeatable, multi-step AI agent processes for tasks like deep research and code review.

A ROS wrapper for Intel RealSense cameras that enables depth and color sensor data to be streamed as ROS topics for robotic applications.