The AI Agent Challenge is a 6-hour online team-based competition about Multi-Agent Systems. This edition brings together students (aged 16 and above) and AI enthusiasts in a single event, all working to solve a real-world case through the collaboration of various AI agents.
All resources are provided at the start, and a training section with simulations is available on the platform to help teams prepare before the competition day.
In this page, you can find some useful information about this initiative and what's in it for your school and request a free promo kit with goodies for your students!
All students participating in the challenge can take part in their dedicated League. After creating a team, make sure they tell us the name of your University or High School in the dedicated widget on the challenge page.
During the Challenge, the points the team earns will contribute to the school’s final score in the Leaderboard. The more students you involve in the game, the greater the chance your school has to win the League.
The University with the most points will win a Reply branded football table game for the communal area, or a financial donation for the Institute.
The High School with the most points will receive a €2,000 donation, plus a coding course designed for your class by Reply experts.
Get ready for the big Challenge on April 16th with your students.
In the Learn & Train section, you’ll find learning pills curated by AI experts to support you in building agents using different frameworks. The platform provides step-by-step guides and resources to help you integrate multiple agents and create collaborative systems.
You can also practice and experiment freely in the dedicated sandbox environment to better understand the game mechanisms.
Important Notes:
Credits: No credits are provided in sandbox mode. Your students can create an OpenRouter account and use the free LLMs available for practice. On challenge day, each team will receive credits as stated in the How It Works section.
Datasets: The sandbox provides the first three datasets for practice. On challenge day, there will be five datasets total—datasets 4-5 will unlock based on the competition rules.
You're free to choose the tools and frameworks that best suit your approach. However, for the competition you must use:
LLMs via API: We will provide you with an API key to access the LLMs available for the challenge
LangFuse: You must integrate LangFuse for tracking purposes following the instructions provided in the Resource management section
Beyond these requirements, you can use any additional libraries, frameworks, or tools to develop your agentic system.
How to set it up:
Follow the detailed guide here: API Guidelines
Download the tracking template from the challenge page: click the "How to Track Your Submission" button
Insert your Langfuse Session ID in the upload modal field for every submission
Important: Langfuse is used to properly track your agent system performance and evaluate it.
The Langfuse host to use is this: http://challenges.reply.com/langfuse
The AI Agent Challenge is a 6-hour competition where your team will build an agentic system to solve a specific problem statement.
Timeline on April 16th:
⏰ Challenge starts at 15.30 PM
💯 Leadearboard frozen at 21.00 until 21.30 PM
✋ Challenge ends at 21.30 PM
🏆 Within 10 working days: podium validation & final results
In your challenge page you’ll access:
Training Dataset:
- Use these datasets to develop and refine your agentic system
- Submit outputs as many times as you want
- Check your score after each submission to track your progress
Evaluation Dataset:
- Submit your final solution only once
- Include both your output and source code (zip file with your agentic system)
- Your final score will be based solely on the evaluation dataset performance
At challenge start: Your team will have access to the first three datasets.
Once your team submits the final evaluation solution for the first three datasets, datasets 4 and 5 will be automatically unlocked.
Resources Provided:
Problem statement
API key for LLM access
Training and evaluation datasets
Token allocation for your team: your team receives a budget in two stages:
Datasets 1-3: $40 in tokens
Datasets 4-5: $120 in tokens (unlocked only after submitting evaluation solutions for datasets 1-3)
Token monitoring dashboard
Tracing dashboard with all your team's Langfuse tracking data
Find all materials, instructions, and tools in the competition page and Train & Learn section.
Check out this link for more detailed information.
The AI Agent Challenge is a 6-hour online team-based competition open to professionals and students (aged 16 or above) worldwide, passionate about AI and Multi-Agent Systems.
All participants compete in the same challenge, but students can also join their dedicated University or High School leagues for additional prizes.
Replyers can take part too, but in a different leaderboard: please follow the communication on TamTamy.
Register online and form a team until 15th April, 23.59 CEST. On April 16th – play from 15:30 to 21:30 CEST.
At the end of the challenge, the Reply Team will review and validate the top scoring submissions from the leaderboard. The first-place team will receive €2,500 for each member, the second-place team €1,500 per member, and the third-place team €1000 per member. The Reply Team’s decision is final regarding competition rules and prize assignment. We’ll send the winners details of how to claim their prize.
Request your kit and promote the AI Agent Challenge in your school.