# Granink

- **Event:** [Meta Global AI Developer Hackathon](https://cerebralvalley.ai/e/meta-ai-developer-hackathon)
- **When:** Oct 26, 12:00 PM – Nov 4, 9:00 PM (EST)
- **Where:** Online
- **Team:** [Bourama DIARRA](https://cerebralvalley.ai/u/Boumi)
- **GitHub:** https://github.com/petitboura
- **Demo video:** https://youtu.be/61y41GBp0sA
- **Gallery:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4449

Project Overview

My project is a multi-model AI platform that brings several large language models (LLMs) together in one place. LLMs are the core of the product: without them, the platform does not exist. Integrating Meta's models is therefore essential, because it lets more users test these models in practice.

I am based in Africa. When I started focusing on my startup, I struggled to pay for subscriptions to services like ChatGPT and Claude, to the point where I had to give them up. I spent a long time looking for an affordable virtual payment card that actually works on these platforms. That is why I built this platform: to make powerful AI models more accessible to Africans by simplifying payments and giving access to several leading LLMs from a single interface.

During the hackathon, improving Classinus with Meta's models was a very good experience for me.

Platform and Repositories

Platform: https://classinus.com

classinus-backend: the platform's server (API, connection to the LLMs, MCP integrations, document and visual content generation).

classinus-frontend: the platform's user interface (multi-model chat, library, configuration).

class-tech: intelligent technical diagnostics for devices (phones first, then PCs), where an LLM guides the user step by step to identify the problem.

Key Features

1. Access to Multiple LLMs: use several large language models, including Meta's models, from a single interface.

2. Isolated Code Execution: users can write and run code in secure, isolated sandboxes.

3. Advanced Search: search academic papers and documents through digital libraries such as OpenAlex and the Internet Archive, search for videos through the YouTube API, and perform web searches through Tavily.

4. MCP Integrations: connect external applications through our built-in MCP integrations or create custom ones.

5. Document Generation and Multimodal Input: generate Word, Excel, PDF, and Markdown documents. The platform also accepts images, audio, video, PDFs, and other document formats as input.

6. Visual Content Generation: integration with the Higgsfield API and its MCP lets users generate images and videos.

7. Smartphone Diagnostics: diagnose smartphones by connecting them to the platform via USB cable.

## More from Meta Global AI Developer Hackathon

- [Anville](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4446)
- [Doc_Coder](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4447)
- [dotbion](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4448)
- [Marksweep (Vanta Point Labs)](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4450)
- [The Resonant Mirror](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4451)
- [rojak-lingo](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4452)

---

Markdown version of https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/4449. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt