Qwen2.5-0.5B-linux-aggressive

🎯 LINUX-optimized | 📦 Aggressive pruning | ⚡ 10% weights pruned

This model is a aggressively pruned version of Qwen/Qwen2.5-0.5B.

Performance Comparison

Category	Original	Pruned	Change
Python	0.0%	0.0%	→
Html	0.0%	0.0%	→
Trivia	100.0%	83.3%	↓ 16.7%
Math	66.7%	66.7%	→
Reasoning	66.7%	66.7%	→
Medical	66.7%	50.0%	↓ 16.7%
Linux	33.3%	16.7% ⭐	↓ 16.7%
Writing	33.3%	33.3%	→

Average: 45.8% → 39.6% (-6.2%)

Linux Retention: 50.0%

Quick Start

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("CompactAI/Qwen2.5-0.5B-linux-aggressive")
tokenizer = AutoTokenizer.from_pretrained("CompactAI/Qwen2.5-0.5B-linux-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/Qwen2.5-0.5B
Specialization	Linux
Prune Mode	Aggressive
Weight Reduction	10% weights pruned

License

This model inherits the license from the base model.