README.md

---
license: apache-2.0
tags:
- pruned
- python
- optimized
- wanda
base_model: Qwen/Qwen2.5-3B-Instruct
pipeline_tag: text-generation
---

# Qwen2.5-3B-Instruct-python-aggressive

> 🎯 **PYTHON-optimized** | 📦 **Aggressive** pruning | ⚡ **35% weights pruned**

This model is a **aggressively pruned** version of [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct).

## Performance Comparison

| Category | Original | Pruned | Change |
|----------|----------|--------|--------|
| **Python** | 92.3% | 84.6% ⭐ | ↓ 7.7% |
| Html | 40.0% | 30.0% | ↓ 10.0% |
| Trivia | 100.0% | 86.7% | ↓ 13.3% |
| Math | 100.0% | 100.0% | → |
| Reasoning | 91.7% | 83.3% | ↓ 8.3% |
| Medical | 64.3% | 35.7% | ↓ 28.6% |
| Linux | 69.2% | 61.5% | ↓ 7.7% |
| Writing | 54.5% | 36.4% | ↓ 18.2% |

**Average**: 76.5% → 64.8% (-11.7%)

**Python Retention**: 91.7%

![Comparison Graph](comparison_graph.png)

## Quick Start

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("CompactAI/Qwen2.5-3B-Instruct-python-aggressive")
tokenizer = AutoTokenizer.from_pretrained("CompactAI/Qwen2.5-3B-Instruct-python-aggressive")

inputs = tokenizer("Your prompt here", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

## Technical Details

| Property | Value |
|----------|-------|
| Base Model | [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) |
| Specialization | Python |
| Prune Mode | Aggressive |
| Weight Reduction | 35% weights pruned |

## License

This model inherits the license from the base model.