Global AI Hackathon Series with Qwen Cloud

About

Build production-ready agents on Qwen — and show the world what an agent can do. The Global AI Hackathon with Qwen Cloud invites professional developers, AI/ML engineers, and builders to design and ship sophisticated multi-agent systems, complex AI workflows, and production-grade applications using Qwen and other flagship models available on Qwen Cloud infrastructure. With advanced reasoning and multimodal capabilities, QwenCloud is built for builders who care about architectural depth and engineering excellence. Pick a track and build something real. Compete for total $70,000+ in cash and cloud credits, get featured on the Qwen Cloud blog, and earn an invite to join the AI Catalyst program. Why Join Build with Qwen Cloud. Access Qwen and other flagship models with advanced reasoning and multimodal understanding — all on Qwen Cloud infrastructure purpose-built for production AI. Get free Qwen Cloud credits to build, deploy, and scale your agent. Win a total of $70,000+ in cash and cloud credits plus blog features, swag, and an invite to the AI Catalyst program for standout teams. Get started Register on Devpost (2 minutes) Sign up for Qwen Cloud. Click here to Sign up for your free trial and request your free hackathon credits via our coupon form. To learn more about the hackathon you can check here.(5 minutes) Join the Qwen Cloud Discord (2 minutes) Pick your track, review the sample projects and reference architecture in the Resources and start building. (30 minutes to first run) Requirements What to Build Build a project using Qwen models available on Qwen Cloud and fall within at least one of these tracks. Track 1: MemoryAgent Build an Agent with persistent memory that autonomously accumulates experience, remembers user preferences, and makes increasingly accurate decisions across multi-turn, cross-session interactions. Participants should focus on:efficient memory storage and retrieval, timely forgetting of outdated information, and recalling critical memories within limited context windows. Track 2: AI Showrunner Leverage video generation capabilities such as Wan / HappyHorse to build an Agent that autonomously handles the entire short drama creation pipeline — from scriptwriting and storyboarding to video generation and editing. Participants should demonstrate the Agent's narrative ability and multimodal orchestration skills while maximizing output quality under a limited token budget. This track has the highest token allowance. Track 3: Agent Society Design a multi-agent collaboration system where multiple Agents with distinct capabilities work together through task division, dialogue, and negotiation to accomplish complex tasks. Participants should showcase: how Agents decompose tasks and assign roles, how they resolve disagreements and execution conflicts, and a measurable efficiency gain over single-agent baselines. Track 4: Autopilot Agent Build an Agent that automates real-world business workflows end-to-end. Scenarios are open-ended — for example, from customer inquiry emails to quote generation, from system alerts to automated remediation, or from resume screening to interview scheduling. Participants should demonstrate the Agent's ability to handle ambiguous inputs, invoke external tools, and incorporate human-in-the-loop checkpoints at critical decision points. Emphasis is on production-readiness over toy demos. Track 5: EdgeAgent Build Qwen-powered physical devices — robots, IoT agents, smart hardware — that perceive via edge sensors, reason via cloud APIs/Skills, and act locally. Participants should demonstrate: robust edge-cloud orchestration under bandwidth/latency constraints, privacy-aware data handling, and graceful degradation in offline/weak-network Scenarios. What to Submit Provide a URL to your code repository for judging and testing. The repository must contain all necessary source code, assets, and instructions required for the project to be functional. The repository must be public and open source by including an open source license file. This license should be detectable and visible at the top of the repository page (in the About section). Include Proof of Alibaba Cloud Deployment: You must demonstrate that the backend is running on Alibaba Cloud with a short recording (separate from your demo) proving your Project’s backend is running on Alibaba Cloud. Please provide this proof in the form of a link to a code file in your code repo that demonstrates use of Alibaba Cloud services and APIs. Include an Architecture Diagram showing a clear visual representation of your system (e.g., how Qwen Cloud connects to your backend, database, and frontend). Include a video (about 3 minutes) that demonstrates your submission functioning. Videos must be uploaded to YouTube, Vimeo, or Facebook Video and made public. Include a text description that should explain the features and functionality of your Project. Identify which Track you are submitting into. Optional: Provide a URL to a published Blog or Social Post showing your journey building with QwenCloud to be eligible for the Blog Post Prize. See more details above for Blog Submission.