---
license: apache-2.0
inference: false 
base_model: Qwen/Qwen2-0.5B-Instruct
base_model_relation: quantized 
tags: [green, p1, llmware-chat, ov]
---

# qwen-2-0.5b-instruct-ov


**qwen2-0.5b-instruct-ov** is an OpenVino int4 quantized version of Qwen2 Instruct 0.5B, providing a very fast, very small inference implementation, optimized for AI PCs using Intel GPU, CPU and NPU.    

[**qwen2-0.5b-instruct**](https://huggingface.co/Qwen/Qwen2-0.5B-Instruct) is the smallest instruct model currently available from Qwen.


### Model Description

- **Developed by:** qwen
- **Quantized by:** llmware
- **Model type:** qwen2-0.5b
- **Parameters:** 0.5 billion
- **Model Parent:** Qwen/Qwen2-0.5B-Instruct
- **Language(s) (NLP):** English  
- **License:** Apache 2.0  
- **Uses:** General purpose  
- **RAG Benchmark Accuracy Score:** NA 
- **Quantization:** int4  
  

## Model Card Contact

[llmware on github](https://www.github.com/llmware-ai/llmware.git)  

[llmware on hf](https://www.huggingface.co/llmware)  

[llmware website](https://www.llmware.ai)