Footer

Your go-to destination for discovering trending open-source projects and uncovering the insights that matter

Legal

  • Terms and conditions
  • Privacy policy

More

  • Pricing
  • Advertise
  • onedump
  • gosnakego
  • Contact us

Social

  • X (Twitter)
  • GitHub
Trendshift owl
What they build speaks louder than words
© 2026 Trendshift, created by Julian Li with ❤️
SignalAdvertise
Trendshift
SignalAdvertise
Sign in
Trendshift owl

Discover trends that matter

DailyWeeklyMonthlyYearly
Live mentionsTopics
RepositoriesDevelopers
InsightsStats

Bifrost
Bifrost
ClawBench
ClawBench
Busbar
Busbar

Tongyi-MAI/OpenGCUA

An open-source stack for generalist computer-use agents, combining code execution, GUI interaction, and cross-device control.

#AI agent
Visit GitHub
HTML
429
33
4 contributors
Apache License 2.0
website
AchievementsView all →
FeaturedAdvertise to 180K+ monthly visitors
Bifrost

Enterprise AI Gateway Built For Scale: Route, Govern, and Secure All Your AI Traffic

ClawBench

a benchmark for personal agents

Busbar

Control what AI can do before it acts

Momentum

How fast Tongyi-MAI/OpenGCUA is gaining stars, and where it lands on Trendshift and GitHub Trending

Activity and trending

Stars, forks and merged pull requests each day or month, lined up with the repository's rank on Trendshift Trending

Oct 15 – Oct 23, 2026

Stars +413 in 9 days

050100150
113

Forks +33 in 9 days

051015
14

Merged PRs 1 in 9 days

01
1

Trendshift Trendingwhere the repository ranked each day it made the top 25

Top 25 in HTML

7146
Oct 15Oct 17Oct 19Oct 21Oct 23
The left and right arrow keys step from day to day; Home and End go to the first and the last

Stars, forks, merged pull requests and Trendshift Trending rank, by day
DayStars gainedForks gainedMerged PRsHTML
Thu, Oct 15, 2026+0+00Not trending
Fri, Oct 16, 2026+17+01Not trending
Sat, Oct 17, 2026+45+20#7
Sun, Oct 18, 2026+34+30#14
Mon, Oct 19, 2026+111+60#6
Tue, Oct 20, 2026+113+140Not trending
Wed, Oct 21, 2026+60+60Not trending
Thu, Oct 22, 2026+33+20Not trending
Fri, Oct 23, 2026+0+00Not trending
Total+413+331
  • 1#1 on the list
  • 7Rank that day
  • Not trending

Mentions and achievements

Where people are talking about Tongyi-MAI/OpenGCUA, and every badge it has earned
  1. Oct 211 post
    • 电脑操作智能体原来被 harness 按死一个节奏:一张截图配一次点击,循环往复。OpenGCUA 把节奏交给模型自己排——什么时候要看屏幕、是一次点一个键还是写段程序带循环跑完、要不要顺手调 app 的 API 或 shell,由模型判断,顺手把模型、harness、轨迹、评测这一整套开放出来。

      Read full on X, posted by @QingQ77

      @QingQ77 · x.com

  2. Oct 202 posts
    • 想研究电脑 Agent 怎么干活,通义这次把不少零件摆出来了。 OpenGCUA 放出了 macOS 控制程序、跨设备 SDK 和评测轨迹。 可以从示例和轨迹看它怎么操作真实设备;这些组件目前还在完善中.

      Read full on X, posted by @Aoyi21

      @Aoyi21 · x.com

    • 阿里通义 MAI-UI 团队开源了 OpenGCUA,一套让 agent 用整台电脑干活的技术栈,协议是 Apache 2.0。 它跟只会看截图点按钮的那种不一样,每一步由模型自己定要不要截图、怎么动手、活怎么排。动手是直接写一段 Python,把无障碍树、应用 API、shell 命令和鼠标键盘操作串在一个程序里,带循环和判断,进度模型自己记在 todo 里。 官方数:在 OSWorld-v2…

      Read full on X, posted by @KinChuan_

      @KinChuan_ · x.com

  3. Oct 191 post
    • 通义把通用 computer use 的全栈开源了 代码执行、GUI 操作和跨设备控制合成一套,官方放出 8 个前沿模型的评测轨迹,附 macOS 14 以上的 harness。

      Read full on X, posted by @aigeeknews

      @aigeeknews · x.com

  4. Oct 181 post
    • 电脑Agent每点一下,都要问模型吗? OpenGCUA示例:Python读控件树,找到Continue并点击,再观察Done。 明确的判断可留在代码里。其macOS harness用mu.emit展示观察,模型下一轮才读取;它不会在代码中即时插入推理。 研究预览源码解读,未实测;OSWorld路径另有区别。

      Read full on X, posted by @anoymous_elangm

      @anoymous_elangm · x.com

  5. Oct 171 post
    • 阿里通义 MAI-UI 团队开源了 OpenGCUA 第一版:computer-use agent 可以把 GUI 操作和无障碍树、API、shell 调用写进同一段程序里执行,截图也由模型按需请求。在 OSWorld-v2 上,GPT-6 Astra、Claude Opus 5 等 8 个模型平均提升 19.6 分,每个任务的步数平均减少 53%。

      Read full on X, posted by @LiYao271158

      @LiYao271158 · x.com

  6. Oct 161 post
    • Verdict: Solid result, will likely replicate — benchmarked across eight models, releases full implementation; real-device harness reveals efficiency gaps closed benchmarks miss.

      Read full on X, posted by @AIEnthusiast26x

      @AIEnthusiast26x · x.com

You might also like

Repositories close to Tongyi-MAI/OpenGCUA
FeaturedAdvertise to 180K+ monthly visitors
ClawBench

a benchmark for personal agents

Bifrost

Enterprise AI Gateway Built For Scale: Route, Govern, and Secure All Your AI Traffic

Busbar

Control what AI can do before it acts

OpenGVLab/ScaleCUA

Python

[ICLR 2026 Oral] ScaleCUA is the open-sourced computer use agents that can operate on cross-platform environments (Windows, macOS, Ubuntu, Android).

1.1k82

xlang-ai/Qwen-CUA

Qwen-CUA: Native Computer Use for (Almost) Everything — a screenshot-driven agent that operates computers with keyboard and mouse, jointly developed by the Qwen Team and XLang Lab.

22713

iFurySt/open-computer-use

Swift

👾 Open Computer Use – Open-Source Alternative to Codex Computer Use

2.6k269

simular-ai/Agent-S

Python

Agent S: an open agentic framework that uses computers like a human

12.7k1.7k

trycua/cua

Rust

Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.

30.7k2.3k

OpenAdaptAI/OmniMCP

Python

Labs: a perceive-plan-act computer-use experiment. Not the OpenAdapt product. The product is openadapt-flow: a compiled program that halts unless an independent check passes.

7718

openai/openai-cua-sample-app

TypeScript

Learn how to use CUA (our Computer Using Agent) via the API on multiple computer environments.

1.9k442

Tencent/UI-Mate

JavaScript

Advancing Foundation GUI Agents with In-Context Demonstrations

11011

vitalops/opendesk

Python

Control 1 or more machines using computer use tools that integrates with your agents

8720