Google Analytics counts our visits, with cookies. No thanks turns it off and removes its cookies; in Europe, the UK, Switzerland and India, or wherever we cannot tell where you are, it stays off unless you accept. Cookie Notice

Advertise
PRICES

Rising AI Repos on GitHub

AI and LLM projects on GitHub ranked by the stars they gained over the last 7 days, not by total stars, and the new projects that took off this week.

Repos tracked60AI and LLM projects
New this week31with stars already
Top gain+5,517wangzan101/Mika

AI repositories

Read (1 h ago); read every 6 hours.
  • wangzan101/Mika New

    An on-device AI agent for Android that operates apps and the web to get things done for you.

    +5,517 stars in 5 days · 5,517 stars · Kotlin

  • modelflare-ai/AnyModelBot New

    🤖 Open-source, local-first alternative to Grok Bot, OpenAI dots and Claude Duet. Your own AI bots on any model: GPT, Claude, Gemini, Grok, DeepSeek, Kimi, GLM and 70+ more…

    +1,553 stars in 7 days · 1,553 stars · TypeScript

  • jackwener/DSH-Android New

    免 ROOT 免 Termux,一个 APK 在安卓手机上跑 DeepSeek Harness。引擎内嵌 App 进程 · 读屏操作 App · 改系统设置 · 定时/条件自动化 · 可选 Shizuku

    +759 stars in 6 days · 759 stars · Kotlin

  • N1nEmAn/sapper-v2 New

    Autonomous pentest legion — AI commander + worker swarm + layered fact-intent graph. 3rd Tencent Intelligent Attack-Defense online champion (634 teams · 8h unattended · 1B tokens…

    +623 stars in 4 days · 623 stars · Python

  • Balanov/uncut New

    The lossless, headless coding agent harness. One 7 MB static binary, four tools, plain stdin/stdout, no silent cuts.

    +575 stars in 1 day · 575 stars · Go

  • wanghao-SunSky/cs329z-chinese New

    斯坦福 Stanford CS329Z Engineering AI Agents 中文教程:面向 Python 开发者的 AI 智能体课程,19 讲与 4 项实践,涵盖 LLM、RAG、工具调用、记忆、多智能体、评测与安全。完整 Markdown、在线与离线 HTML。Chinese course notes.

    +459 stars in 5 days · 459 stars · HTML

  • jmanico/grim-prompter New

    Threat modeling for AI coding agents: point Claude Code, Codex or any agent at your repo and get an evidence-traced threat model. 36 architectures, 29 methods, MIT.

    +349 stars in 6 days · 349 stars

  • davemessew/AgentFlux New

    Agentic AI platform for visual workflows and autonomous agents. Run it with Docker on your own computer or server.

    +248 stars in 7 days · 248 stars · TypeScript

  • aa1143/NamiAgent New

    零基础入门 AI Agent 的基础教程。不只是教程,也是一个能在本地跑起来的 Agent:读写文件、执行命令、MCP、长期记忆,自带网页端看清每一步。从 100 行最小循环开始,一课加一个能力,最后拆解 Claude Code、Codex 等主流 Agent

    +237 stars in 4 days · 237 stars · Python

  • muaaz9329/cheap-ag New

    50x your usage with cheap-ag. Opus/Astra plans, a dirt-cheap model writes every line, and tests decide what ships. Your $20 plan stops hitting limits. One MCP server for Claude…

    +231 stars in 5 days · 231 stars · TypeScript

  • tavern-labs-ai/persona-ai-sdk New

    Persona AI SDK: default personas, durable character memory, and Bedrock/OpenAI adapters in one package.

    +198 stars in 3 days · 198 stars · TypeScript

  • SAOS-3624/dsh-see-now New

    让 AI 实时看见你的屏幕:按变化量筛选帧、多帧拼进一张图,静止画面几乎不花 token

    +190 stars in 2 days · 190 stars · JavaScript

How this page works, and questions

Our server searches GitHub every 6 hours for active repositories tagged llm, ai-agents or generative-ai, and for new ones created in the last two weeks, and notes each one’s star count every day.

“Stars gained” is the difference between today’s count and the oldest count we hold from the last 7 days; a repository younger than a week gained all its stars in that time. Forks and archived projects are left out.

Stars measure attention, not quality or safety. Read a project’s code and licence before you use it.

How is this different from GitHub Trending?

GitHub Trending ranks a single day across every topic. This list keeps AI and LLM projects only and ranks them by a week of growth, so a steady climber is not pushed out by a one-day spike.

Why is a famous project missing?

Only projects tagged with the topics above are searched, and only those pushed to in the last 30 days.

Figures come from the public sources named on each page, read by our server on a schedule, and may be a few minutes or hours old. Nothing here is investment advice; see the risk disclaimer.