Trending AI projects

GPT4Free is a community-driven aggregator that provides a unified interface and API for accessing various LLMs and media-generation models, allowing developers to switch between providers easily.

DirectorDesk is a cross‑platform desktop tool for building simple 3‑D stages, choreographing character actions, and designing camera moves. It exports reference videos (including depth maps) that can be fed to AI video‑generation models, and it offers a natural‑language AI assistant and support for external agents to automate edits.

Fizgig is an open‑source desktop‑and‑web app for training LoRAs, slider/edit LoRAs, and full fine‑tunes on a range of image, video and music diffusion models (Flux Klein 9B, Krea 2, MiniMax H3, Qwen Image 2.1, SDXL, etc.). It works on modest GPUs (8 GB+) or on rented RunPod GPUs, offering tools for data prep, automatic dataset curation, visual epoch browsing, and block‑wise LoRA repair, all under an Apache 2.0 licence.

Sim is an open‑source platform for building, deploying, and monitoring AI agents and workflows. It offers a web UI (and a macOS desktop client) with visual flow editing, chat‑first agent creation, and built‑in tables, files, and knowledge bases. The platform connects to 270+ integrations and any major LLM, supports local models via Ollama/vLLM, and runs on a stack of Next.js, Bun, PostgreSQL, and various supporting libraries. Self‑hosting is done via an interactive `npx sim-setup` wizard that provisions Docker‑Compose services, generates secrets, and manages the deployment. Licensed Apache‑2.0.

该项目提供一套工具无关的手绘画风提示词配方（21套编号画风），附带渲染脚本，可将风格、内容、比例等参数组合成完整提示词或带参考图的JSON调用包，供AI Agent在图像生成时使用，并包含严格的验收与生产限制流程。

NVIDIA Model Optimizer is an open‑source Python library for compressing and speeding up large AI models (LLMs, VLMs, diffusion). It supports PTQ, quant‑aware training, pruning, distillation, speculative decoding, and sparsity, works with Hugging Face, PyTorch, or ONNX inputs, and exports directly to TensorRT‑LLM, vLLM, SGLang, etc. Install via `pip install nvidia-modelopt[all]` or use NVIDIA containers. The repo includes docs, examples, and pre‑quantized checkpoints.

A DeepSeek Harness plugin that adds OAuth‑based login and UI management for subscription‑based LLM providers (ChatGPT (Codex), Claude, Grok, Copilot, Google Antigravity). It integrates their live model catalogs into DSH’s model picker, shows quota usage, and registers tools for web search, image generation, and video generation, all without needing API keys.

A curated collection of 159 installable agent skills that let AI assistants reason from the perspective of famous thinkers, leaders, and creators — from Confucius to Buffett — turning each figure into a complete thinking framework rather than a quote book.

A ComfyUI plugin that provides a visual timeline and orchestration tools for the MiniMax H3 video model, enabling seamless long-form video generation, character replacement, and audio-driven lip sync.

A DSH plugin that provides native access to OpenCode Zen models without requiring API keys or accounts by mimicking official CLI traffic.

A collection of 50 MIT-licensed, fully built AI SaaS templates (Stripe billing, Google OAuth, MuAPI models, Vercel-ready) that anyone can fork, rebrand, and sell — with high-margin credit-based pricing math laid out in the README.

A unified command-line tool and TUI for managing configurations, API providers, MCP servers, and session history across multiple AI coding assistants like Claude Code, Codex, and Gemini.

OpenEdit is an open‑source CLI skill that lets AI coding agents (Claude Code, Codex, Gemini) edit videos by describing the desired changes. It handles transcription, layout, rendering in a headless browser and outputs an MP4, supporting subtitles, cuts, motion graphics and AI‑generated media.

An autonomous improvement engine for AI coding agents that uses a loop of modification, mechanical verification, and git-based rollbacks to optimize code and systems.

A curated directory of Claude skills — reusable add‑on tools for the Claude AI assistant — organized by category, with links to each skill's repository and a brief description.

Open Lovable is a demo app that lets you chat with an AI to generate React apps instantly. It requires a Firecrawl API key, one of four AI provider keys, and a sandbox provider (Vercel or E2B). Setup involves cloning, installing, configuring environment variables, and running the dev server. It's MIT-licensed.

A local logging reverse-proxy and dashboard that lets you see exactly what prompts, tool calls, and tokens your AI coding agents are sending to LLM providers.

A collection of optional plugins for Claude Code that use function hooks to surface live status lines, guard against destructive commands, show what subagents are doing, and add panes and games. It installs only the mods you choose, making Claude Code more transparent and controllable.

A multimodal data layer and visual debugger for physical AI, allowing developers to log, query, and visualize multi-rate sensor data in real-time.

Zoo Code is a VS Code extension that adds an AI‑powered assistant capable of generating, refactoring, debugging, documenting, and automating code via natural‑language prompts. It offers several work modes, fast semantic search, safe command guarding, and integrates with many LLM providers.

UniForge3D is an AI system that turns a few unposed photos of an object into a complete 3D model. It jointly reconstructs observed geometry and generates missing parts in a shared latent space, using a view‑selection step to preserve details. The repo provides a Conda‑based install, a simple `infer.py` script, MIT‑licensed code, and pretrained weights (subject to their original licenses).

VoidAccess is an open‑source Python tool (MIT‑licensed) that lets security researchers ask a natural‑language question and receive a dark‑web‑sourced threat‑intel report. It scrapes dozens of Tor search engines, code hosts and public feeds, filters unsafe content, enriches entities with LLM‑generated summaries, builds a searchable relationship graph, and exports results in formats like STIX, MISP, Sigma, YARA, CSV or JSON. It can be run locally via `pip install voidaccess` (with optional Tor and LLM configuration) or used as a hosted service at voidaccess.pro.

PiDeck is an open‑source Electron‑based desktop workbench that wraps the pi coding‑assistant CLI (and DeepSeek Harness) to let developers manage multiple projects and AI coding‑agent sessions with integrated file, Git, terminal, and model‑configuration UI.

A terminal-based AI coding agent that understands codebases, performs multi-file edits, and executes commands with a focus on speed and remote SSH efficiency.

The MCP TypeScript SDK provides TypeScript libraries to build Model Context Protocol (MCP) servers and clients, enabling standardized, type‑safe delivery of tools, resources, and prompts to LLMs. It runs on Node, Bun, and Deno, offers middleware for popular web frameworks, and includes docs, examples, and migration guides.

TensorRT‑LLM is NVIDIA’s open‑source library that compiles large language (and visual‑generation) models into ultra‑fast TensorRT inference engines. It provides custom CUDA kernels, quantization, sparse attention, speculative decoding, and multi‑GPU parallelism, delivering 2‑10× higher token throughput while keeping accuracy. The Python‑friendly runtime integrates with Triton, Kubernetes, and Jetson, and includes docs and examples for models like GPT‑OSS‑120B, Llama‑4, DeepSeek‑R1, and diffusion models.

An MCP server that enables AI assistants to search, create, and update content in Jira and Confluence across Cloud and Server/Data Center deployments.

goofish-cli is a Python CLI that automates Xianyu (Chinese second‑hand marketplace) operations—publishing items, managing listings, sending/receiving messages, searching, etc. It outputs data in JSON/YAML/table formats, includes rate‑limiting and risk‑guard checks, and registers as an MCP server so AI agents (Claude Code, Cursor, OpenClaw) can call its commands as tools. Five Claude skill bundles are shipped for zero‑shot‑free agent use. The package is pip‑installable, open‑source under Apache 2.0, and comes with unit tests and documentation.

An AI-powered content creation platform that uses multi-agent collaboration to automate trending topic discovery, writing, layout design, and publishing for social media platforms like WeChat.

A lightweight, dependency-free C/C++ tensor library for machine learning designed for high efficiency and cross-platform portability.

MAKO is a community‑maintained Vulkan layer (with an optional Decky Loader UI) that adds AI‑driven frame generation, up‑scaling, and bundled post‑process shaders to games on Steam Deck and Linux. It works with the proprietary Lossless Scaling models (user‑supplied) and supports per‑game profiles, adaptive frame rates, HDR, and both 32‑/64‑bit Vulkan processes.

The Context Ontology Accelerator is an AWS‑native open‑source platform that builds a semantic knowledge graph from an organization’s data sources, validates context against business rules, and serves it to AI agents via SPARQL or a custom MCP interface. It combines ontology reasoning (HermiT/ELK), virtual KG federation (Ontop), and AWS CDK‑provisioned infrastructure, targeting enterprises that need reliable, rule‑checked context for LLM‑driven agents.

An open-source trading engine and MCP server for prediction markets that identifies and trades probability-based arbitrage opportunities across Kalshi, Polymarket, and Solana.

A cross-device shared memory layer for AI agents that uses encrypted cloud drive synchronization to maintain a continuous semantic memory across different platforms and devices.

A Python CLI tool that uses Chrome DevTools Protocol to log in to BOSS 直聘, fetch job listings with clear salary data, optionally scrape detailed descriptions, analyse skill frequencies, and output JSON/CSV plus a résumé‑optimisation prompt.

An autonomous, collaborative research pipeline that turns a research idea into a conference-ready academic paper with real citations and sandbox experiments.

An integrated management system for Xianyu sellers that automates account handling, AI-driven customer replies, and digital product delivery.

InferenceX™ is an Apache‑2.0 open‑source platform that continuously benchmarks LLM inference performance across many hardware platforms and software stacks. It offers end‑to‑end, networking, and kernel‑level benchmark suites, a live public dashboard, and tooling for running reproducible tests on‑premise or in the cloud.

An agent skill that transforms complex academic papers into structured, evidence-based Obsidian notes, automating the ingestion process for academic wikis.

skillshare is a Go‑based CLI (and optional desktop app) that stores, versions, audits and syncs AI‑coding “skills”, agents, rules and plugins across tools like Claude Code, Codex, Pi, OpenCode, etc. It offers a unified source of truth, security audits, multi‑tool sync, and a visual dashboard.

seekdb is an open‑source, MySQL‑compatible database that adds vector and full‑text search, async indexing, and copy‑on‑write sandboxes—features tailored for AI agents that need fast continuous writes, immediate retrieval, and safe branching. It works embedded, via Docker, or as a cloud instance, integrates with LangChain/LlamaIndex/Dify, and is Apache‑2.0 licensed.

A set of 36 Claude‑Skill packages that turn Anthropic’s Claude LLM into a domain‑expert for major GRC frameworks (ISO 27001, SOC 2, FedRAMP, GDPR, HIPAA, etc.). Install the `.skill` files, mention a relevant term, and Claude automatically provides gap analyses, policy drafts, control mappings and technical guidance, all with up‑to‑date regulatory citations.

A cross-platform desktop app for orchestrating multiple AI coding agents and projects in parallel, featuring automated playbooks and Git worktree integration.

An open-source integration layer that centralizes tool configurations, authentication, and policies for AI agents via the Model Context Protocol (MCP).

A unified OpenAI-compatible API gateway providing access to high-performance multimodal foundation models for text, image, and video generation.

A personal, open-source collection of LLM knowledge and interview questions, organized as structured markdown notes covering everything from fundamentals to advanced topics like RLHF and RAG, with companion hands-on projects for practical learning.

A comprehensive automation tool for batch-registering ChatGPT accounts and performing Codex OAuth authorization, featuring multiple browser drivers, email integration, and a local WebUI.

A data-driven field guide to AI engineering roles, skills, and interviews, built from 6,964 real job descriptions, interview experiences, and practitioner stories. It covers role analysis, interview prep (51 companies), transition paths from five backgrounds, portfolio guidance, raw job market data, curated resources, webinars, and a course/community.