Llama-3.2-1B-Instruct - Renesas X5H

Introduction

This repository contains Llama-3.2-1B-Instruct model, optimized for Renesas X5H platform for text inference.

• Model Architecture: Llama 3.2-1B is an auto-regressive language model that uses an optimized transformer architecture.
• Source Model: meta-llama/Llama-3.2-1B-Instruct

Performance

The following performance metrics were measured with a prompt.

Model	Precision	Device	Response Rate (tokens/sec)
Llama-3.2-1B-Instruct	F16	X5H - Single Cluster NPX	16.7 tokens/sec

Prerequisites

To run model, you need:

1. Renesas X5H Board
2. llama-runner CLI:
3. Hugging Face CLI: For downloading the model.

Deployment

Copy the binary and model to one single folder

<PATH_ON_BOARD>
├── llama-runner
├── Llama-3.2-1B-Instruct-f16.gguf

Inference

./llama-runner "prompt"