---
title: Qwen3-1.7B-Qnn-AOT-SM8650
canonical_url: "https://www.modelscope.cn/models/mllmTeam/Qwen3-1.7B-Qnn-AOT-SM8650"
md_url: "https://www.modelscope.cn/models/mllmTeam/Qwen3-1.7B-Qnn-AOT-SM8650.md"
repository: mllmTeam/Qwen3-1.7B-Qnn-AOT-SM8650
last_updated: 2026-02-08
parameters: 2.1B
tensor_type:
  - F32
  - U8
  - I32
  - I8
  - U16
library_name:
  - safetensors
downloads: 86
stars: 0
---

# Qwen3-1.7B-Qnn-AOT-SM8650

> Qwen3-1.7B-Qnn-AOT-SM8650 - mllmTeam 在 ModelScope 开源的模型。Qwen3-1.7B W4A16 QNN Model for SM8650

- **Repository**: mllmTeam/Qwen3-1.7B-Qnn-AOT-SM8650
- **Parameters**: 2.1B
- **Downloads**: 86
- **Stars**: 0
- **Last updated**: 2026-02-08

Source: https://www.modelscope.cn/models/mllmTeam/Qwen3-1.7B-Qnn-AOT-SM8650

---

# Qwen3-1.7B W4A16 QNN Model for SM8650

- **model.safetensors**: Post-training quantization (PTQ) calibrated model
- **model.mllm**: Model converted from `model.safetensors` by `mllm-convertor`
- **qwen3-1.7B-lpbq-sha.bin**: QNN model exported via AOT (Ahead-of-Time compilation), the final executable model