Trending AI projects

A fast and memory-efficient implementation of the attention mechanism for transformer models, reducing GPU memory IO to increase speed and efficiency.

A scalable reinforcement learning algorithm that uses world models to master diverse control tasks with a fixed set of hyperparameters.

A pair of Claude Code skills that enforce a spec-driven workflow: /spec designs a feature spec through clarifying questions, and /spec-impl implements it step by step only after you manually approve the spec. This prevents silent LLM improvisation, makes decisions explicit, and keeps every change reviewable.

Vosk is an offline open-source speech recognition toolkit that provides continuous transcription for over 20 languages across devices ranging from Raspberry Pi to large clusters.

A human-centric framework for building and manage real-life AI and ML systems, streamlining the transition from prototyping to production deployment.

A tool and field guide designed to identify and remove generic, AI-generated design patterns (AI slop) from web projects.

An ultra-low-latency, self-hosted speech-to-text pipeline that enables real-time transcription and translation using advanced simultaneous speech research and multiple model backends.

A framework for self-improving LLMs that evolves system prompts into structured playbooks through a modular process of generation, reflection, and curation.

ParadeDB is a PostgreSQL extension that adds fast full‑text, BM25, and vector similarity search (including hybrid queries) plus filtering, faceting, and OLAP‑style aggregates—all inside a single SQL‑compatible database. Built with Rust (pgrx, Tantivy, DataFusion), it can be run locally via Docker or as a managed cloud service, and integrates with popular ORMs, AI agent tools, and PaaS platforms.

This is a comprehensive, free guidebook for the Hermes Agent framework, detailing how to use its self-improving learning loops, multi-layer memory, and multi-agent orchestration features.

Kubeflow is an open‑source, Kubernetes‑native platform that bundles modular AI/ML tools—training, pipelines, notebooks, serving, etc.—into a portable, scalable stack for the entire model lifecycle.

An automation tool for Kubernetes that manages the installation and lifecycle of NVIDIA GPU drivers, runtimes, and monitoring tools as containers.

A Python wrapper for the Ollama AI inference server, offering chat, generation, streaming, async calls, model management, embeddings, and a decision‑making System One API. Install via `pip install ollama` and point the client at a local or cloud Ollama instance.

HeliosGen is an open‑source desktop app (macOS released, Windows/Linux pending) that provides a visual, node‑based canvas for building and running AI image and video generation pipelines. It works locally, sending requests to the kie.ai API (or optionally a Codex CLI) and stores all data on the user’s machine. Features include multi‑model support, reference images, parallel/sequential execution, and a real‑time history. The app is built with Tauri 2 (Rust) and a Next.js/React front‑end, uses SQLite for local persistence, and is MIT‑licensed.

An AI‑driven creative‑writing assistant built as Claude‑compatible agents and skills. Muse coordinates brainstorming, drafting, critique, editing, and a growing knowledge base, letting writers maintain voice and continuity across a story.

A high-performance, native Rust CLI agent for AI-assisted coding that prioritizes low resource usage, secure tool execution, and predictable terminal interaction.

OneCLI is an open-source platform that allows teams to deploy and manage secure, sandboxed AI agents for every employee while centralizing credential management and security policies.

autoMate is a locally hosted service that gives AI chat clients persistent personal storage (notes, files, reminders, long‑term memory) and a toolbox of real‑world actions (shell, browser, SaaS APIs). It talks to clients via an MCP‑over‑HTTP bridge, works with OpenClaw, Claude Desktop, Cursor, Cline, etc., and can also be used through its own web chat, Chrome extension, Docker container, or binary.

Amadeus is an open-source, real-time multimodal desktop agent that combines voice conversation, a speech-synced animated character, and task execution by specialized providers into a single interface. It aims to solve the fragmentation of separate voice assistants, desktop characters, and execution agents by keeping talk, embodiment, action, and user control in one coherent loop.

A plugin and server implementation that exposes Blockbench 3D modeling tools to AI agents via the Model Context Protocol, enabling automated model creation and editing.

A voice-controlled interface for Claude Code that allows users to brainstorm designs, drive automated builds, and monitor multiple AI coding sessions via voice.

Switchyard is an open‑source NVIDIA NeMo library that routes each LLM request to the most appropriate model from a user‑defined pool, letting you blend cheap fast models with larger, higher‑quality ones. It offers multiple routing algorithms (Auto, Task‑based classifier, Execution‑based stage router, Composite, etc.), can be used via gateways like OpenRouter, LiteLLM, or NeMo Relay, run as a local proxy, or embedded directly in a Rust/Python harness, and is pre‑1.0 Apache‑2.0 software for cost‑/latency‑aware AI agents.

SwarmUI is an open‑source, web‑based UI for running AI image, video, and audio generation models (Stable Diffusion, Flux, etc.). It offers a beginner‑friendly Generate tab, an advanced Comfy‑style workflow editor, multi‑GPU “swarm” support, and extensible plugins. Installers exist for Windows, Linux, macOS (Apple‑silicon), and Docker, with cloud‑GPU templates for Runpod and Vast.ai. The project is in an “Almost‑Release” beta (v0.9.8) and is MIT‑licensed, though it can auto‑install GPL/AGPL components.

LetsFG is an AI-integrated travel search and booking engine that provides real-time, transparent pricing for flights and hotels, allowing AI agents to perform bookings directly without the price inflation of traditional travel sites.

A high-performance Rust-based engine that detects duplicate code, dead code, and complexity hotspots across 224+ languages, featuring AI-ready integration via an MCP server.

LightX2V is an open‑source inference framework for fast image and video generation. It supports many models (MiniMax‑H3, LTX‑2, HunyuanVideo‑1.5, Wan‑2.x, Qwen‑Image, etc.) and offers speed‑up tricks like 4‑step distilled LoRAs, FP8/NVFP4 quantisation, tensor‑parallelism, and disaggregated deployment on a wide range of accelerators. Users interact via a single `LightX2VPipeline` class, choosing a model, variant and a JSON config, then call `generate` with a task (e.g., t2av, t2v, i2i). Reported benchmarks show up to ~4× speedup over Diffusers on H100 GPUs. The project provides Docker images, a macOS app, online demo, and extensive docs and examples.

Awesome Agents is a curated, open-source directory of tools, frameworks, and platforms designed to help developers build, evaluate, and manage AI agents.

An open-source, community-curated list of use cases, demos, and repositories built with Jev, TypeSafe's decision model. It includes popularity metrics, cost/latency reports, model limitations, and ideas, all in one place.

A hands-on educational repository that teaches .NET developers how to build and deploy Generative AI applications using practical code samples, agent frameworks, and cloud or local model hosting.

A library of AI‑agent “skills” that let tools like Cursor or Claude Code perform App‑Store‑Optimization and mobile‑app growth tasks (keyword research, audit, metadata writing, competitor analysis, ad‑campaign planning, etc.) using real data from the Appeeky API. Install via `npx skills add …` or as a Claude plugin, then ask natural‑language queries; the router skill routes them to the right specialist skill and returns actionable reports.

An open agentic coding system that uses multi-agent systems to transform natural language and research papers into production-ready code.

Dapr is an open‑source CNCF‑graduated runtime that runs as a side‑car and provides durable workflows, secure service‑to‑service calls, state management, pub/sub, actors, and a Conversation API for LLMs. It lets you build reliable AI agents and distributed applications in any language, with built‑in security, observability, and the ability to run on Kubernetes, cloud VMs, edge devices, or locally.

A native Mac application and CLI that accelerates local LLM inference on Apple Silicon using Multi-Token Prediction (MTP) for faster decoding speeds.

A set of reliable PyTorch implementations of reinforcement learning algorithms designed to provide stable baselines for research and industry.

A portable, one-click installer for ComfyUI on Windows that includes a management dashboard for environments, packages, and hardware configurations.

MetaHarness is a CLI and web studio that creates a repo‑specific AI‑agent harness (custom CLI, skills, memory, and policy) which can run on multiple host agents and be published as an npm package.

A comprehensive, community-maintained directory and marketplace for DeepSeek Harness plugins that features automated discovery, strict quality verification, and an integrated in-app store.

Jacobian is an MCP server that provides AI agents with a searchable, executable library of precise mathematical operations, enabling them to perform reliable calculations and compose complex solutions.

An AI-powered social media diagnostic tool that analyzes account performance, deconstructs viral content, and generates data-driven sales scripts and content strategies for creators.

A scikit-learn compatible tabular foundation model that enables zero-shot classification and regression on mixed-type tabular datasets using in-context learning.

BabelDOC is an open‑source Python tool that translates PDF scientific papers using LLMs (OpenAI‑compatible models) and outputs bilingual PDFs. It offers a CLI, a Python API, extensive PDF‑handling options, glossary support, offline asset packaging, and can be run locally or via a hosted beta service.

A locally‑run Python tool that statistically fingerprints an official LLM API and compares it with a third‑party API to detect possible tampering, using repeated prompts and low token consumption.

SEO Machine is a Claude Code workspace that automates SEO‑focused content creation. It offers slash‑commands for research, writing, analysis, optimization, and WordPress publishing, backed by agents that evaluate keyword density, readability, intent, CRO, and more. Integrated with GA4, Search Console and DataForSEO, it turns real‑time performance data into actionable content briefs and scores, helping marketers produce data‑driven blog posts and landing pages.

Kaku is a macOS‑only terminal built on WezTerm that ships with ready‑made fonts, themes, and shortcuts, and adds AI‑assistant features such as command suggestions, natural‑language‑to‑command conversion, and an integrated chat panel. Install via Homebrew or DMG, configure AI services yourself, and enjoy remote‑access and Lua‑based customization.

海鸥 2.0 是一个为 Claude Code 安装“暴躁安全研究员”人格的配置包，包含大量对话示例和术语映射，支持跨平台安装/卸载，自动备份原配置。

A git-native AI agent framework where identity, memory, and tools are stored as version-controlled files, allowing agents to be branched and diffed like code.

ComfyUI‑LTXVideo adds custom nodes and example workflows to run Lightricks’ LTX‑2 video‑generation model inside the ComfyUI visual‑programming UI, providing text‑to‑video, LoRA‑based controls, HDR conversion, dubbing, up‑scaling and high‑resolution tiled sampling.

Qiushi‑Skill is a plug‑in library that adds nine philosophically‑derived reasoning methods (e.g., contradiction analysis, investigation‑first, mass‑line) to AI agents such as Claude Code, Cursor, and Codex. Installed via `npx qiushi-skill` or the Claude Code marketplace, it injects a lightweight kernel that enforces the principle “seek truth from facts” and routes tasks to the appropriate skill. Each skill provides trigger rules, step‑by‑step actions, output templates, hard constraints, and hand‑off specs, plus reference texts and optional sub‑agents for investigation or self‑criticism. The project aims to make agents more evidence‑driven, self‑critical, and strategically focused.