LFM2.5-1.2B-Thinking-linux-aggressive

🎯 LINUX-optimized | 📦 Aggressive pruning | ⚡ 30% weights pruned

This model is a aggressively pruned version of LiquidAI/LFM2.5-1.2B-Thinking.

Performance Comparison

Category	Original	Pruned	Change
Python	25.0%	5.0%	↓ 20.0%
Html	45.0%	5.0%	↓ 40.0%
Trivia	100.0%	72.5%	↓ 27.5%
Math	100.0%	77.5%	↓ 22.5%
Reasoning	97.5%	85.0%	↓ 12.5%
Medical	45.0%	32.5%	↓ 12.5%
Linux	75.0%	42.5% ⭐	↓ 32.5%
Writing	80.0%	45.0%	↓ 35.0%

Average: 70.9% → 45.6% (-25.3%)

Linux Retention: 56.7%

Quick Start

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("CompactAI/LFM2.5-1.2B-Thinking-linux-aggressive")
tokenizer = AutoTokenizer.from_pretrained("CompactAI/LFM2.5-1.2B-Thinking-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))