Trending AI projects

An open-source deception framework that deploys realistic, LLM-powered decoys to observe attacker behavior and generate security evidence.

Ladybug is an embeddable, high-performance graph database designed for complex analytical workloads and large-scale data retrieval, featuring native vector indexing and full-text search.

A visual drag-and-drop builder for creating and deploying AI agents and workflows without writing code.

A library of portable, structured investment judgment systems that allow AI agents to apply the specific strategies and risk-management rules of legendary investors.

A framework that converts local LLMs and VLMs into structured decision models by extracting probabilities from logits during prefill, bypassing token-by-token decoding for lower latency.

3D Gen Studio is an open‑source desktop/web app that lets you build 3‑D assets with AI. It combines a Kanban board or node‑graph UI, tight integration with ComfyUI workflows, and support for external mesh‑generation APIs. Built‑in mesh editing includes UV unwrapping, auto‑retopo, auto‑rig, sculpting, painting and game‑ready checks. Projects are stored locally (SQLite/LowDB) and can be shared over LAN. Install via npm for the frontend and a Python backend for Blender‑based tools.

A skill installation platform for Claude Code, OpenCode, and Codex that provides a curated collection of academic and scientific AI capabilities.

A cognitive-driven autonomous security agent for authorized penetration testing that uses causal graph reasoning and a Planner-Executor-Observer architecture to ensure traceable, evidence-backed security research.

Ludwig is a declarative deep learning framework that enables users to train, fine-tune, and deploy LLMs, multimodal, and tabular AI models using YAML configuration files instead of boilerplate Python code.

A tool that distills the viral strategies of Xiaohongshu and Douyin influencers into AI-ready guides and detailed reports to help creators replicate successful content patterns.

A metasearch library that aggregates results from multiple web search services and provides content extraction tools for text, images, videos, and news.

StatsPAI is an open‑source Python library that replicates many Stata and R econometric commands (regression, IV, DiD, RD, synthetic control, matching, etc.) with a Stata‑style API, provides agent‑ready metadata, built‑in datasets, and utilities for exporting tables and interactive plotting.

An open-source SDK for AI agent safety that blocks prompt injections, redacts sensitive data, and scans repositories for AI-targeted threats.

Litho (deepwiki‑rs) is a Rust‑based tool that automatically generates C4‑style architecture documentation from a codebase using LLMs. It scans source files, extracts structure and comments, runs AI agents to infer architecture, creates Mermaid diagrams, and outputs a synced wiki. Supports many languages, can integrate external docs, and includes companion tools (Litho Book viewer, Mermaid Fixer). Designed for CI/CD integration and aims to eliminate manual documentation upkeep.

A Gradio-based web interface for the browser-use library that enables users to interact with AI browser agents using a variety of LLMs.

An open-source AI video creation workbench that uses an infinite canvas and a natural-language assistant to manage the pipeline from script to final video.

AHE is an observability-driven framework that automatically evolves the system components surrounding a coding agent—such as prompts and tools—to improve task performance without retraining the underlying model.

MOSS‑VL is an open‑weight, 11 B parameter video‑language model series that supports real‑time interaction on live video streams (MOSS‑VL‑Realtime) as well as offline high‑quality video QA (MOSS‑VL‑Instruct) and a base checkpoint for further training. It uses a cross‑attention architecture with timestamps and 3‑D rotary embeddings, offers interruptible streaming, proactive silence, and dynamic correction, and provides quantized checkpoints that run on a single 24 GB GPU. The repo includes ready‑to‑run inference scripts, a FastAPI/WebSocket server, SGLang‑based serving back‑ends, fine‑tuning scripts (full‑parameter and LoRA), and extensive benchmarking results.

Microsoft / skills is a genuine open‑source toolbox that equips AI coding agents (e.g., GitHub Copilot) with ready‑made, Azure‑specific “skills” – code snippets, templates, and MCP server scaffolding – so the agents can generate correct Azure SDK calls, manage Foundry resources, and follow governance patterns. Install via `npx skills add microsoft/skills`, pick only the needed skills (≈175 across Python, .NET, TypeScript, Java, Rust), and use the provided `AGENTS.md` and MCP configs to build custom agents for backend, frontend, infrastructure, or planning tasks.

An open-source interpretability platform for exploring and analyzing neural network features, providing tools for automated interpretation, steering, and circuit visualization.

A unified AI SDK for Laravel that provides a consistent API to interact with providers like OpenAI, Anthropic, and Gemini for agents, images, and audio.

SAHI is a Python library that adds tiled (sliced) inference to existing object‑detection models, enabling accurate detection of small objects in very large images or videos. It supports many popular detectors, offers a CLI and FiftyOne integration, and includes COCO‑dataset utilities.

Chisle is a plugin for AI coding agents that reduces token costs by compressing tool outputs and enforcing a concise, efficient coding style for the model's responses.

HunyuanOCR‑1.5 is a lightweight OCR‑focused vision‑language model that unifies document parsing, table extraction, and text‑image translation. It speeds up long‑output decoding with DFlash speculative decoding, offers a unified vLLM/Transformer inference environment, and can be run on laptops via llama.cpp. The repo provides scripts for data packing, supervised fine‑tuning, DFlash draft training, and ready‑to‑use inference servers.

An automated comic translator that uses LLMs, OCR, and inpainting to translate comics and manga from multiple languages while preserving the original visual layout.

A collection of Claude Code skills and playbooks that automate B2B cold email infrastructure, lead sourcing, and copywriting based on patterns from 1,000+ campaigns.

这是一个用于机器人竞赛（RoboMaster）的电控教学教程仓库，包含从入门到进阶的系统化课程、代码、视频和幻灯片，面向新队员和传承困难的队伍，遵循 CC BY-NC 协议，仅限于队内交流使用。

A historical archive for the OWASP Top 10 for Large Language Model Applications, providing access to legacy documentation and redirecting users to the current active security project.

A curated list of apps and tools that work with OpenRouter, a unified API for accessing hundreds of AI models from major providers. The directory helps users discover applications that integrate with OpenRouter, from coding agents to chat clients, observability tools, and more.

A visual speech recognition tool that converts silent lip movements captured by a webcam into text pasted directly into any application.

An Obsidian plugin that uses local embeddings to automatically surface semantically related notes, helping users discover connections in large vaults without manual tagging.

AgentFS is a SQLite-based filesystem designed for AI agents that provides persistent storage, key-value state management, and a full audit trail of tool calls in a single portable file.

A GitHub guide for GLM 5.3 abliterated/uncensored models, including jailbreak prompts, API access via RouterHub, and evaluation tips.

NVIDIA RTX Streamline is an open‑source SDK that abstracts NVIDIA’s AI‑based graphics enhancements (DLSS, DLSS‑G, Reflex, NIS) into a single integration layer for Windows games. It provides pre‑built signed DLL plugins, a unified API, build scripts, debugging tools, and sample code, allowing developers to add super‑resolution, frame generation, latency reduction, and image scaling without writing vendor‑specific code.

A locally‑run Excalidraw canvas that AI coding agents can control via CLI, MCP, or REST. Supports element‑level CRUD, headless screenshots, deterministic .excalidraw exports, Mermaid import, snapshots, and Obsidian‑compatible files – all without external API keys.

Meridian is a Node.js‑based local server that bridges coding‑assistant clients (Pi, OpenCode, etc.) to Anthropic’s Claude and Google’s Antigravity models, offering unified APIs, caching, session management, and a macOS dashboard.

A collection of free, browser-run Google Colab and Kaggle notebooks for generative AI tools — covering video upscaling, text-to-image, text-to-video, expressive text-to-speech, voice cloning, music generation, and talking-head avatars. It removes hardware and setup barriers so anyone can run advanced models on free cloud GPUs.

A local, privacy-focused tool that uses OCR to index the text inside your screenshots, allowing you to search and retrieve them instantly by content.

An automated evaluation platform that uses vision models to test, score, and report on the quality of AI-generated applications across web, Android, and iOS platforms.

ARDY is an autoregressive diffusion model for real-time, interactive human motion generation that supports text-to-motion synthesis and precise kinematic constraints like waypoints and keyframes.

Claude of Tanks is a browser‑based, Three.js‑powered tank combat simulator featuring 220 procedurally generated vehicles, 33 destructible battlefields, 13 game modes, and WebRTC multiplayer, all built with a Claude/Codex AI‑assisted pipeline.

An open-source, self-hosted gateway that connects phone lines and messaging apps to AI agents, allowing for private, AI-powered receptionists and customer communication.

BidMaster Pro is an AI‑driven desktop‑plus‑server platform that automates the full tender‑to‑bid process. It parses tender documents, generates bid drafts with LLMs, runs 21 compliance checks, and formats the final document. Built on FastAPI, React‑Electron, LangGraph agents, a plug‑in Skill engine, and a RAG vector store (ChromaDB), it supports multiple LLM providers (DeepSeek, OpenAI, Alibaba Tongyi, local Ollama) and OCR via MinerU. Docker‑based deployment and extensible Python Skills make it suitable for companies that need fast, compliant bid preparation.

FastWAM is a high-performance World Action Model for robotics that enables fast action prediction by optionally skipping test-time future imagination, significantly reducing inference latency.

A JetBrains‑maintained plugin that supplies AI coding assistants with version‑aware guidelines for writing modern Go (up to Go 1.27). It integrates with Junie, Claude Code, Codex, Cursor, and generic `skills.sh` agents, letting them automatically prefer newer language features and idioms.

A privacy-focused, on-device voice dictation tool for Linux that converts speech to typed text in any active application using local STT engines.

A community-curated library of 6,582 text prompts for ByteDance's Seedance 2.0 video generator, with 6 featured examples, sample videos, author credits, a companion web gallery, and a CC BY 4.0 license.

CodeGraphContext is a CLI toolkit and MCP server that indexes local code into a graph database, enabling AI agents and developers to perform structural code analysis and trace complex relationships across repositories.