EXAONE-4.0-1.2B
Collection
Collection of pruned models based on EXAONE-4.0-1.2B
•
16 items
•
Updated
🎯 HTML-optimized | 📦 Aggressive pruning | ⚡ 35% weights pruned
This model is a aggressively pruned version of LGAI-EXAONE/EXAONE-4.0-1.2B.
| Category | Original | Pruned | Change |
|---|---|---|---|
| Python | 76.9% | 23.1% | ↓ 53.8% |
| Html | 20.0% | 20.0% ⭐ | → |
| Trivia | 86.7% | 46.7% | ↓ 40.0% |
| Math | 80.0% | 66.7% | ↓ 13.3% |
| Reasoning | 75.0% | 25.0% | ↓ 50.0% |
| Medical | 42.9% | 14.3% | ↓ 28.6% |
| Linux | 23.1% | 30.8% | ↑ 7.7% |
| Writing | 54.5% | 0.0% | ↓ 54.5% |
Average: 57.4% → 28.3% (-29.1%)
Html Retention: 100.0%
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("CompactAI/EXAONE-4.0-1.2B-html-aggressive")
tokenizer = AutoTokenizer.from_pretrained("CompactAI/EXAONE-4.0-1.2B-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))
| Property | Value |
|---|---|
| Base Model | LGAI-EXAONE/EXAONE-4.0-1.2B |
| Specialization | Html |
| Prune Mode | Aggressive |
| Weight Reduction | 35% weights pruned |
This model inherits the license from the base model.
Base model
LGAI-EXAONE/EXAONE-4.0-1.2B