Qwen3-1.7B-Q4 - Optimized 4-bit ONNX for Mobile
Copyright (c) 2026 Edison dos Santos (Edison2ST)

This product includes software developed by:
- Alibaba Cloud (Qwen team): Original Qwen3-1.7B model architecture and weights.
- Microsoft Corporation: ONNX Runtime engine and Microsoft Olive optimization framework.
- Edison dos Santos: ONNX conversion, Q4 quantization, and tensor alignment optimization for mid-range mobile hardware.

Part of the Ghost Assistant project: https://github.com/Edison2ST/GhostAssistant