Challenge Rules
Complete guide to the 2026 New Year Challenge
2026 New Year Challenge Overview
The 2026 New Year Challenge is a global, collaborative hackathon jointly hosted by five leading AI projects. Running for 10 days, from Jan 8 to Jan 18, 2026 (PT), the event offers a total prize pool of $3,000.
Each project hosts its own independent track, covering diverse aspects of modern AI agent development, from memory management and automation to external integrations and deployment.
You can participate in multiple tracks simultaneously, and each track will be evaluated separately based on its own scoring.
Tip: Click a project tag below to jump to its dedicated rules page.
MemU PR Hackathon
😸Event Overview
The MemU PR Hackathon is a 10-day online open-source challenge open to developers worldwide.
During the event, participants contribute directly to the MemU open-source repository by submitting real, mergeable Pull Requests (PRs). Together, we aim to build next-generation, memory-native agent infrastructure.
You can:
- Fix bugs
- Build new features
- Create integrations
- Write or improve documentation and tutorials
- Explore and ship your own ideas
Each merged PR earns points. At the end of the hackathon, top contributors will receive cash rewards and limited-time community honors.
🗓️Timeline
Jan 8 – Jan 18, 2026 (Pacific Time)
Online · Global · Open-source
💰Scoring, Prizes & Participation Terms
🔹Scoring Rules
- Only PRs submitted during the hackathon period are eligible for scoring
- Hackathon period: Jan 8, 00:00 PT – Jan 18, 00:00 PT
A PR must:
- Be submitted during the event window
- And be successfully merged to earn points
All PRs will be reviewed by the MemU team. PR reviews will be completed within 48 hours of submission.
Each merged PR earns points based on its track or official evaluation:
Points are accumulated and used for final ranking and reward redemption.
🔹Prizes & Recognition
Participants in the MemU PR Hackathon will be rewarded through three types of incentives:
1. Points-based Cash Prizes
The top three contributors will receive $300 / $200 / $100 respectively, based on their points ranking within this track.
(The first-place prize winner must complete at least one task worth 3 points or more.)
2. Badge Recognition
Every merged PR earns a badge, with higher difficulty tasks granting higher-tier badges:
3. MemU Platform Credits
All points accumulated in this hackathon can be converted into MemU platform credits at a predefined conversion rate.
📏Tracks & Rules
The challenge is organized into multiple tracks, covering system-level features, integrations, documentation, and testing. All submissions will be primarily evaluated based on usability.
1. Challenges & Task Types
The MemU PR Hackathon features two types of challenges:
🔹 Open Challenges
If you have a feature or idea you want to build, you can participate through Open Challenges.
- Open an Issue in the GitHub repository describing the feature or improvement you want to implement, and add the label: #2026NewYearChallenge
- The MemU team will review the issue and assign a score based on its scope and complexity within 10 hours: 1 / 3 / 5 points
- After the score is confirmed, implement the solution and submit a PR
- Once the PR is merged, you earn the assigned points
- New features or capabilities
- Integrations not covered by official tasks
- Experimental or exploratory engineering work
As long as the submission is usable, mergeable, and valuable, it is eligible for scoring.
🔹 Official Tasks
A set of predefined challenges designed by the MemU team, covering different difficulty levels and point values.
Some tasks already have issues created. You can directly claim issues labeled #2026NewYearChallenge and complete your submission within the specified timeframe.
2. Challenge Rules (Official Tracks)
🔹 General Rules
- All submissions must be made via PRs
- Each PR must belong to one track only
- Individuals and teams are welcome to participate
PRs must:
- Not break existing core functionality
- Follow MemU's architectural and design principles
- Include at least one runnable or verifiable outcome
🏆5-Point Tracks
(High Difficulty · High Impact)
Designed for contributors delivering complete solutions, covering design, implementation, and real usage scenarios.
Track A: Agent Framework Plugins / Adapters
Build plugins or adapters for popular agent or workflow frameworks (such as n8n, AutoGPT, Dify, LangGraph, LlamaIndex), enabling developers to integrate MemU into existing agent projects with minimal effort.
- A usable adapter or plugin
- Clear example code or demo workflows
- Alignment with the target framework's usage conventions
- Usability
Track B: Flexible Workflow (Advanced Memory Orchestration)
Enable MemU to invoke external task orchestration systems at specific memory stages, creating flexible and composable memory workflows.
- A runnable integration solution
- Example code and documentation
- At least one complete, runnable end-to-end example
- Usability
🌟3-Point Tracks
(Medium Difficulty · Official Tasks)
Official issue-based tasks will be assigned. Participants can claim issues from the GitHub repository and submit PRs during the challenge period.
Track C: Model Provider Extensions
Add support for additional LLM or embedding providers, such as Claude, Gemini, OpenRouter, Grok, or other local or open-source models. MemU currently supports OpenAI by default. If embedding models are required, they must be included in the implementation.
- Complete model integration
- No breaking changes to existing model logic
- Documentation
- Basic test cases
- Usability
Track D: External Tool / Database Integration
Enable MemU to synchronize memory with external databases or storage systems, supporting import, export, and backup workflows.
- A runnable data sync, import, or export solution
- Clear read/write interface documentation
- Defined synchronization logic
- Usability
- Data integrity
Track E: Bugfixes & Stability Improvements
Identify and fix real, reproducible issues in MemU.
- Clear problem description and reproduction steps
- Corresponding fix implementation
- Necessary validation or test evidence
- Usability
Track F: SDK Implementation
Develop an SDK for MemU in a programming language of your choice (Python, Go, Java, JavaScript, etc.), enabling developers to interact with MemU programmatically. The SDK should include clear usage examples and comprehensive documentation to ensure easy adoption.
- A complete SDK implementation in at least one language
- Clear and concise API documentation
- Example code demonstrating key functionality
- Optional: basic test cases for core features
- Usability and clarity of the SDK
- Quality and completeness of documentation
- Correctness and reliability of SDK functions
Track G: Sealos-based memU Use Case
Build a practical memU use case using Sealos Devbox as the development and deployment environment. Participants should demonstrate how memU's memory capabilities improve real-world agents or applications, with a runnable demo or clearly documented setup.
- A memU-based use case developed using Sealos (Devbox recommended)
- Clear explanation of the scenario and how memory is used
- Runnable demo, deployment link, or screenshots (if publicly accessible)
- Basic setup or deployment instructions
- Practicality of the use case
- Overall usability and demonstration quality
💪1-Point Tracks
(Low Barrier · High Participation)
Track H: Documentation Improvements
Improve MemU documentation to help new users get started faster, including from-scratch tutorials, scenario-based examples, common pitfalls and best practices. Participants must also publish related content on external platforms (blogs or social media) and clearly reference the official MemU repository.
- Meaningful documentation improvements
- Clear explanations of core concepts
- Consistent and explicit repository references
- Proof of external publication
- Usability
Track I: Testing (Unit / End-to-End)
Write unit tests or end-to-end tests for MemU to improve overall stability.
- Test code
- A brief explanation in the PR description
- Necessary validation steps
- Usability
🌍Open Source & Compliance Statement
To maintain a healthy and sustainable open-source community, all participants must follow these principles:
All submitted code must:
- Comply with the MemU repository's open-source license
- Not introduce third-party code with incompatible licenses
Contributions must be:
- Original work
- Clearly compliant with open-source redistribution terms
The MemU team reserves the right to:
- Make the final decision on whether a PR is merged
- Disqualify submissions that violate project guidelines or policies
✍️Additional Notes
- Individuals or teams may submit multiple PRs; points are cumulative
- In case of a tie, the following factors may be considered: code quality, practical usability, community impact
- The MemU team reserves the right to reasonably interpret and adjust the rules if necessary