LFM2.5-1.2B-Instruct-html-aggressive

🎯 HTML-optimized | 📦 Aggressive pruning | ⚡ 35% weights pruned

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

Performance Comparison

Category	Original	Pruned	Change
Python	50.0%	50.0%	→
Html	83.3%	66.7% ⭐	↓ 16.7%
Trivia	91.7%	0.0%	↓ 91.7%
Math	100.0%	8.3%	↓ 91.7%
Reasoning	66.7%	8.3%	↓ 58.3%
Medical	75.0%	33.3%	↓ 41.7%
Linux	16.7%	16.7%	→
Writing	33.3%	0.0%	↓ 33.3%

Average: 64.6% → 22.9% (-41.7%)

Html Retention: 80.0%

Quick Start

from transformers import AutoModelForCausalLM, AutoTokenizer

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