LFM2.5-1.2B-Thinking
Collection
Pruned models based on LiquidAI/LFM2.5-1.2B-Thinking
β’
16 items
β’
Updated
π― HTML-optimized | π¦ Safe pruning | β‘ 1% weights pruned
This model is a conservatively pruned version of LiquidAI/LFM2.5-1.2B-Thinking.
| Category | Original | Pruned | Change |
|---|---|---|---|
| Python | 45.0% | 52.5% | β 7.5% |
| Html | 67.5% | 70.0% β | β 2.5% |
| Trivia | 92.5% | 87.5% | β 5.0% |
| Math | 97.5% | 97.5% | β |
| Reasoning | 97.5% | 97.5% | β |
| Medical | 65.0% | 62.5% | β 2.5% |
| Linux | 47.5% | 47.5% | β |
| Writing | 92.5% | 90.0% | β 2.5% |
Average: 75.6% β 75.6% (+0.0%)
Html Retention: 103.7%
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("CompactAI/LFM2.5-1.2B-Thinking-html-safe")
tokenizer = AutoTokenizer.from_pretrained("CompactAI/LFM2.5-1.2B-Thinking-html-safe")
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 | LiquidAI/LFM2.5-1.2B-Thinking |
| Specialization | Html |
| Prune Mode | Safe |
| Weight Reduction | 1% weights pruned |
This model inherits the license from the base model.
Base model
LiquidAI/LFM2.5-1.2B-Base