Advantech-EIOT/qualcomm_qwen_qwen2.5-7B_Instruct

qualcomm_Qwen2.5-7B-Instruct

• —Model creator: Qwen
• —Original model: Qwen2.5-7B-Instruct

This repository contains the Qwen2.5-7B-Instruct model optimized for Qualcomm hardware using Qualcomm® AI Engine Direct (QNN). It is designed for high-performance inference on edge devices powered by Qualcomm Snapdragon platforms, enabling efficient on-device AI capabilities with low latency and reduced power consumption.

Model Details

• —Developed by: Advantech-EIOT / Alibaba Cloud
• —Architecture: Qwen2.5
• —Task: Text Generation (Chat/Instruction)
• —Precision: Quantized (w4a16) for NPU optimization
• —Target Device: Quantized and optimized specifically for Snapdragon® X Elite
• —Optimization: Qualcomm® AI Stack / QNN SDK

Hardware Compatibility

This model is highly optimized for Advantech Edge AI platforms powered by Qualcomm processors:

• —Windows: Snapdragon® X Elite (e.g. Snapdragon® based Microsoft Surface Pro)
• —Linux: Dragonwing® Platforms (e.g. Dragonwing® IQ-9075)

Limitations and Disclaimer

Qwen2.5 is a powerful language model but may exhibit hallucinations or generate inaccurate information.

• —Accuracy: Users should validate outputs for critical applications.
• —Usage: Please refer to the Apache 2.0 License for usage restrictions and acceptable use policies.
• —Edge Optimization: Inference performance may vary depending on the specific hardware configuration and thermal constraints of the edge device.