Trending AI projects

A comprehensive, hands-on educational resource that teaches users how to build, train, and fine-tune large language models from scratch using practical code examples and video tutorials.

Solace Agent Mesh is an agent development and runtime platform that hosts AI agents, runs their tools, and coordinates them over an event broker. The repository currently provides Agent Mesh Desktop (a free app with the `sam` CLI) and examples, not the source code yet. Production requires a commercial license via Kubernetes or Agent Mesh Cloud.

Awesome Jev is a curated, source-backed list of projects built with Jev, TypeSafe AI's System One model for fast, typed, probabilistic decisions. It covers official resources, provider integrations, framework adapters, SDKs, agents, and research, with links and notes on limitations.

An observable and replaceable enterprise RAG infrastructure that provides a controllable pipeline for document parsing, chunking, and retrieval with built-in regression testing.

An AI-powered job hunting assistant that automates resume submission and HR communication on Boss Zhipin using LLMs to act as a candidate's digital twin.

A collection of academic writing best practices and an automated LaTeX checking tool designed to help researchers avoid common errors in conference paper submissions.

A curated learning roadmap that consolidates 218 resources across OpenAI, Anthropic, Google, MCP, harnesses, evals, and production agent systems into a phased path, with reading lists, runnable Python code, and build exercises to help you become a world-class Agentic Engineer.

A multi-agent system that converts PyTorch programs into verified Triton kernels and iteratively optimizes their performance using hardware profiling and LLM-guided bottleneck diagnosis.

A high-performance, Rust-powered data engine designed to process multimodal AI workloads, allowing users to handle images, audio, and video alongside structured data at scale.

gImageReader is a graphical interface for the Tesseract OCR engine that allows users to easily extract text from images, PDFs, and scanned documents.

DeepStream is an NVIDIA SDK that provides a GStreamer-based framework for building high-performance, multi-stream AI video and image analytics pipelines on NVIDIA GPUs.

从故事创作、角色资产管理、分镜设计，到 AI 图片与视频生成，打造完整的开源 AI 漫剧与短剧创作工作台。支持创作助手、无限画布、本地部署与多模型接入。

A hands-on tutorial repo that teaches engineers to build AI agents from scratch using plain LLM APIs, tools, and a loop — no frameworks until you understand the primitives. It progresses from a first API call to production-grade agents, covering patterns, testing, loop engineering, and framework comparisons.

An Android application that turns a mobile device into a Text-to-Speech server, enabling other apps to access speech synthesis services.

WorldSculpt is a research codebase that converts multi‑view RGB video (with instance masks and 3D boxes) into a single compositional mesh for scenes containing hundreds of objects. It fine‑tunes the Pixal3D single‑object prior, provides a new UE‑MeshyScene benchmark, releases checkpoints and data on Hugging Face, and includes a community ComfyUI integration.

AURA is an open‑source SRE‑agent platform that lets you run AI‑driven incident‑investigation assistants inside your own infrastructure. It supports multiple LLM providers, configurable guardrails, OpenTelemetry tracing, and dozens of integrations (Kubernetes, Prometheus, PagerDuty, etc.). Deployable as a CLI, Docker image, Helm chart, or embedded Rust library.

MASFactory is a Python framework that lets you build, visualise, and run LLM‑based multi‑agent workflows as graphs. It supports natural‑language‑first design (Vibe Graphing), code‑first declarative/imperative authoring, and visual drag‑and‑drop editing via a VS Code extension, all feeding a unified runtime with live tracing and human‑in‑the‑loop support.

An MCP server that enables AI models to generate over 26 types of professional charts and diagrams using the AntV visualization library.

This repository is a curated list of datasets for applying deep learning to satellite and aerial imagery, including Sentinel-1 and Sentinel-2 products, remote sensing hubs, and specialized collections for tasks like flood mapping, crop classification, cloud detection, and more.

Phistory archives and visualises system prompts of many AI coding‑assistant CLIs, letting users compare how agents’ hidden instructions change over time.

An open standard for shared agent learning that allows AI agents to store and retrieve structured knowledge to avoid repeating mistakes.

Re-debugging Solved Failures? No Way. 500+ Evidence-graded fix lessons, One MCP Call-->https://misakanet.org

A CLI tool that converts long-text documents into structured, searchable knowledge graphs with source-traceable entities and relations based on the LLM Wiki concept.

StackQL is an open-source tool that allows users and AI agents to manage and query cloud resources and SaaS APIs using SQL semantics instead of provider-specific APIs.

The official Python SDK for Deepgram, enabling developers to integrate automated speech recognition, text-to-speech, and language understanding into their applications.

An industrial deep learning platform providing a unified framework for training and inference, automatic parallelism, and support for scientific computing.

A C++ toolbox for solving nonlinear optimal control problems, enabling real-time Model Predictive Control (MPC) for robotic systems with limited onboard computation.

A high-performance C++ toolkit for the simulation of articulated mechanisms, used for biomechanical and robotic systems through a multibody dynamics engine.

Free Crypto News API is a free, no-key API that aggregates news from 300+ cryptocurrency sources and offers 450+ endpoints for news, market data, AI analysis, and more, including an MCP server for AI agents.

An agent-human collaboration platform that uses the Model Context Protocol (MCP) to synchronize AI agent tasks with a visual workspace and remote machine execution.

A comprehensive firmware for ESP32-CAM modules that enables motion-triggered or continuous video recording to SD cards, live streaming, and integration with home automation systems.

CodeGraff is a lightweight, dependency-free AI coding agent that runs in your terminal to automate file editing, command execution, and complex development workflows.

The NVIDIA Container Toolkit is a set of utilities that enables containers to automatically detect and use NVIDIA GPUs, simplifying the deployment of GPU-accelerated applications.

A collection of practical workflows and guides for integrating AI agents into software development, focusing on automating code reviews, security scans, and design verification.

Claude Code: The Complete Guide is a practical, PDF-based ebook from the Orange Book series that teaches you to master Anthropic's terminal AI coding tool — from installation to advanced workflows, real-world projects, and exercises. It's written by a no-code developer who built a #1 App Store app, so it's grounded in real agentic experience.

An Android application that enables local, on-device inference of small language models (SLMs) using the llama.cpp engine.

An open-source GRC engineering toolkit for Claude that automates evidence collection, gap assessment, and remediation across multiple compliance frameworks.

OpenGUI is a framework that lets AI agents control real Android phones by interpreting screenshots and issuing UI actions. It provides WorkBuddy and DeepSeek Harness plugins for easy chat‑based use, a full optional backend for advanced task planning, and a small Android client that performs taps, swipes, typing, etc. Ideal for AI‑driven testing, automation, and content creation on mobile devices.

A collection of 55 portable workflows and security-focused sandbox tools that enable consistent, safe, and specialized task execution across multiple AI coding agents.

A self-hosted AI app builder: one HTTP request spawns an isolated container, runs a coding agent against your prompt, and returns a live preview URL — with your own models, data, and infrastructure. The idea: instead of sending your code to a hosted builder like Bolt or v0, you own the whole stack. sandboxd is intentionally small — Go + Docker + Traefik + SQLite, no Kubernetes — and favors operational simplicity: idle containers sleep and wake automatically, so many apps fit on one machine. Two ways to use it: a REST API for embedding into products, or a no-code web console (a pure API client) for interactive building. Beyond AI generation, it can launch 80+ curated open-source apps as one-click containers. Not for everyone: the README says skip it if you just need a couple of containers (a Docker script suffices). It's aimed at scenarios like AI app-builder products, agent platforms, per-user preview environments, or team multi-app hosting. Monetization: MIT-licensed core, optional managed cloud and BYOC services.

A local proxy that routes requests from various AI coding tools through your existing Claude or ChatGPT subscriptions, eliminating the need for separate per-token API billing.

TurboDiffusion is a video generation acceleration framework that speeds up end-to-end diffusion generation by 100-200x using attention acceleration and timestep distillation.

Judge0 is an open-source online code execution system that compiles and runs untrusted code in a sandbox, supporting 90+ languages via a simple HTTP JSON API. It is designed for building applications like AI agents, e-learning platforms, online judges, and IDEs that need secure, scalable code execution.

SwiftData Pro is an agent skill that teaches AI coding assistants like Claude Code, Codex, and Gemini to write better SwiftData code, focusing on the mistakes LLMs commonly make with models, queries, predicates, indexes, migrations, and iCloud sync. It was created by a SwiftData book author and installs into multiple coding assistants to improve code correctness and efficiency.

A framework that enables multimodal AI models to view a computer screen and drive mouse and keyboard actions to reach objectives. Supports GPT-4o, Claude 3, Gemini, and others, plus OCR and Set-of-Mark modes for better clicking.

DocLayout-YOLO is a real-time model for detecting document layout elements (text, tables, figures, etc.) that uses a large synthetic dataset and a global-to-local perception module to handle diverse document styles and element scales.

A comprehensive, offline-first personal web application that provides a persistent, multi-modal environment for long-term emotional interaction with AI models.

A Python‑based, markdown‑driven framework that records threat‑hunting investigations using a four‑step LOCK template, makes them searchable, and adds AI‑ready context. It provides a CLI, a validation skill (GATES), and an optional local MCP server so LLM assistants can read past hunts, generate hypotheses, and decide whether findings should become production detections.