Llama Impact Hackathon
Address real-world challenges using Llama 3.2 models!
Important Info
To submit your project, you need an account on the lablab.ai platform and you need to be enrolled to this Hackathon.
The email you use to register for the lu.ma event must be the same as the one you used to register on the lablab.ai platform.
Hackathon Challenge
Join this hackathon to tackle real-world challenges using Meta’s Llama 3.2
This event is part of the global Llama 3.1 Impact Grant series, giving participants the opportunity to indicate their participation on their Llama 3.1 Impact Grant application. By doing so, you’ll be eligible to compete in an additional specialized "Regional" track within the grant program. If you’re interested in applying for the Llama 3.1 Impact Grants, we recommend submitting your application after participating in the event.
For more details on the Llama 3.1 Impact Grants, visit: Llama Impact Grants.
1. Navigating Public Services
📌 Objective: Build AI tools that enhance public access to government services by creating intuitive, Llama-based solutions.
Challenge: Use Llama to develop chatbots or virtual assistants that can navigate publicly available government information, helping users with tasks like reporting issues, filing taxes, or understanding city benefits. Additionally, Llama can assist government agencies in improving processes like permitting and grant writing for organizations.
Example: Albert
A Llama 2-based AI developed for French public administration, guiding users through administrative tasks and providing information.
2. Reducing Barriers for Llama Developers
🛠️ Objective: Simplify the development and deployment process for Llama, empowering more developers to utilize this technology effectively.
Challenge: Create tools or methods that reduce common hurdles in Llama development, such as data preparation, evaluation, and prompt engineering. By addressing these barriers, participants can make Llama more accessible and versatile for a broader developer community.
Example: Llama Development Toolkit
Tools for automating data labeling and setting evaluation benchmarks specifically for Llama models, making development faster and easier.
3. Expanding Low-Resource Languages
🌐 Objective: Broaden Llama's accessibility by incorporating low-resource languages and delivering culturally sensitive solutions.
Challenge: Enable Llama models to support languages with limited technological resources, ensuring that AI delivery respects cultural nuances and relevance. This challenge focuses on expanding Llama’s language capabilities and making AI tools accessible to a wider, more diverse audience.
Examples:
AIM Intelligence’s Llama-Suho – Tailors Llama for Korean-specific challenges, addressing cultural sensitivities and hate speech.
Traversaal.ai – Provides bilingual resources in English and Urdu, with upcoming expansions to Arabic for localized educational content.
Prizes
💰 $15,000 cash prize pool for the winners!
⭐ Unlock a $100,000 Opportunity with Meta's Llama 3.1 Impact Grant
Our Partners
We Partnered with leading tech companies to provide you with the tools and resources needed to build exceptional projects. Using any of these technologies in your project gives you access to exclusive resources and rewards from our partners, including specialized support, and additional prizes!
These are some of our partners:
TBA
To participate in our hackathon, you need to be enrolled to the Llama Impact Hackathon on lablab.ai website. This applies to both online and onsite participants.
Event Schedule
Day 1 (November 8th)
6:00 PM: Doors open (Registration + Dinner)
6:00 PM: Hackathon Kick-off (On-site and Online)
6:05 PM: Meta AI Opening words
6:15 PM: lablab.ai Opening words + Hackathon walkthrough
6:30 PM: Panel 1 - Value of Open Source AI
7:00 PM: Panel 2 - Open Source AI as a Building Block
7:30 PM: Networking
9:00 PM: Doors close
Day 2 (November 9th)
10:00 AM: Doors open (Breakfast + coffee is served)
10:45 AM: Overview of the Hackathon
12:00 PM: Lunch
6:00 PM: Dinner
9:00 PM: Doors close
Day 3 (November 10th)
9:00 AM: Doors open (Breakfast + coffee is served)
4:00 PM: End of submissions
4:00 PM: Judging starts
5:30 PM: Finalists live demo
6:30 PM: Winners announcement