SMILE LLM

Phi-2 Model Deployment on Raspberry Pi 5

This project involves deploying the Phi-2 large language model (LLM) on a Raspberry Pi 5 with 8GB RAM, utilizing llama.cpp for optimized local inference.

SMILE LLM represents our commitment to making AI-powered education accessible even in resource-constrained environments. By running language models locally on affordable hardware like the Raspberry Pi 5, we can bring intelligent tutoring and educational support to communities without reliable internet connectivity.

Why Local LLM?

• No internet required: Once deployed, the model runs entirely offline
• Privacy-first: Student data never leaves the device
• Cost-effective: Raspberry Pi 5 costs a fraction of cloud-based solutions
• Scalable: Each classroom can have its own AI assistant

Technical Overview

Using llama.cpp, we optimize the Phi-2 model for ARM architecture, enabling real-time inference on the Raspberry Pi 5's limited hardware. This approach demonstrates that meaningful AI-powered educational tools don't require expensive infrastructure.

The SMILE LLM project builds on our legacy of using the SMILE platform to transform education in underserved communities, now enhanced with the power of artificial intelligence.