Makatia/microsoft_Phi-3-mini-4k-instruct_onnx_rpi

Phi-3 Mini 4K Instruct (ONNX, Raspberry Pi)

Efficient, quantized ONNX export of Phi-3 Mini 4K Instruct, optimized for local inference on resource-constrained devices like the Raspberry Pi.

---

📦 Model Overview

• Archive: microsoft_Phi-3-mini-4k-instruct_onnx_rpi.tar.gz
• Format: Quantized ONNX
• Recommended for: ONNX Runtime, Kleidi AI, and similar frameworks

Unpack the archive to access the ONNX model and configuration files.

---

🚀 Quick Start

Install ONNX Runtime if you haven't already:

pip install onnxruntime

Run inference with the model:

import onnxruntime as ort

session = ort.InferenceSession("phi-3-mini-4k-instruct.onnx")
# Replace with your input and inference code

---

🛠️ Features

• Optimized for Raspberry Pi: Low memory footprint and fast inference
• Quantized: Reduced model size for edge deployment
• Open Source: MIT licensed

---

📋 Credits

• Base model: Microsoft
• ONNX export & quantization: Optimum, ONNX Runtime

---

Maintainer: Makatia