Tongyi-MAI GitHub avatar

Qwen-UI-Agent

Tongyi-MAI

A bilingual website template for the Qwen-UI-Agent technical report, featuring an interactive carousel, benchmark results, demos, and release materials.

Stars

6

7-day growth

No data

Forks

1

Open issues

0

License

No data

Last updated

2026-07-30

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It provides a ready-to-deploy, fully bilingual site designed specifically for AI agent technical reports, with easy customization and GitHub Pages integration.

Who it is for

  • AI researchers preparing technical reports
  • Open-source contributors building demo websites
  • Technical writers creating bilingual documentation
  • Developers showcasing model capabilities

Use cases

  • Presenting an agent's interactive capabilities via carousel and demos
  • Displaying benchmark results with publisher logos
  • Hosting bilingual technical report websites on GitHub Pages
  • Demonstrating real-device mobile and computer-use workflows

Strengths

  • Bilingual (English/Chinese) support from the start
  • Single-file content editing for carousel, benchmarks, and demos
  • Built-in GitHub Pages deployment workflow
  • Includes interactive carousel and embedded video demos

Considerations

  • Tailored specifically to Qwen-UI-Agent, not a generic template
  • Requires npm setup and local development environment
  • Some benchmark figures are omitted until final values are frozen
  • External reference embeds are explicitly labeled as temporary samples

README quick start

Qwen-UI-Agent — Technical Report Website

A concise, application-first website template distilled from the current Qwen-UI-Agent LaTeX draft. Its narrative runs from an interactive real-world capability carousel to benchmark results, broader general and agentic capabilities, playable demos, and release materials, with a complete English / 中文 interface.

Local preview

npm install
npm run dev

Open http://localhost:3000.

Production checks

npm run lint
npm test
npm run build:pages

npm run build:pages creates the static GitHub Pages output in out/. The project is configured for a root Pages site and ships public/.nojekyll. .github/workflows/deploy-pages.yml builds and publishes this output whenever the main branch is pushed.

Editing guide

  • Edit all bilingual copy, capability-carousel items, benchmark cards, broader-capability tables, and video links in app/siteContent.ts.
  • APPLICATIONS controls the six visual slides in the interactive “what it can do” carousel. Each card intentionally contains only a visual, label, title, and short description.
  • Capability visuals support a CSS-drawn mobile scene, a muted looping local video, a three-frame animated browser sequence, and regular image crops.
  • MODEL_ORGANIZATIONS maps each benchmark entry to its visible publisher label and local asset in public/brand-logos/.
  • GENERAL_CAPABILITY_GROUPS is rendered inside Performance as the “Broader Capabilities” subsection; it is not a standalone page section.
  • Desktop performance cards use a three-column layout and collapse responsively on narrower screens.
  • The five real-device mobile workflows are hosted directly from public/demos/source/; edit their bilingual titles and instructions in DEMO_VIDEOS.
  • The two full-length Computer Use workflows are also hosted from public/demos/source/ as web-optimized 720p H.264 videos. Their complete bilingual task instructions live in DEMO_VIDEOS.
  • The Demo explorer uses five domains: real-device mobile, computer use, cross-device GUI use, mobile use with Deep Research, and proactive service. The cross-device domain embeds the two official Bilibili workflows for phone-to-PC receipt processing and mobile-search / PC-summary work. The two proactive workflows are locally hosted portrait videos with bilingual instru

Related repositories

Similar projects matched by category, topics, and programming language.

yetone
Featured
yetone GitHub avatar

kill-ai-slop

Kill AI Slop is a multilingual field guide and agent skill that catalogues 33 common AI-generated design tells with interactive before-after demos, and provides a tool to scan and fix web projects.

AI & Machine LearningAI Agents
805
react
Featured
react GitHub avatar

react

React is a JavaScript library for building user interfaces with declarative, component-based, and cross-platform design.

Web DevelopmentFrontend Frameworks
246,741
MoonshotAI
Featured
MoonshotAI GitHub avatar

Kimi-K3

Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.

AI & Machine LearningAI Agents
3,348