lfm25-encoder-agnews-smoke
LiquidAI/LFM2.5-Encoder-350M fine-tuned for single-label text classification on fancyzhx/ag_news.
- Labels (4):
World,Sports,Business,Sci/Tech - Date: 2026-07-28 16:37 UTC
[!NOTE] This model uses a custom classification head (mean pooling over a backbone without a native sequence-classification class), so loading requires
trust_remote_code=True. vLLM serving requires a standard architecture.
Evaluation
| Metric | Value |
|---|---|
| accuracy | 0.7575 |
| f1_macro | 0.7568 |
Usage
from transformers import AutoModelForSequenceClassification, AutoTokenizer
model = AutoModelForSequenceClassification.from_pretrained("davanstrien/lfm25-encoder-agnews-smoke", trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained("davanstrien/lfm25-encoder-agnews-smoke", trust_remote_code=True)
inputs = tokenizer("your text here", return_tensors="pt", truncation=True)
print(model.config.id2label[model(**inputs).logits.argmax().item()])
Reproduction
Produced on Hugging Face Jobs (gpu) with the train-classifier.py recipe from uv-scripts. Run it yourself:
hf jobs uv run --flavor gpu --secrets HF_TOKEN \
https://huggingface.co/datasets/uv-scripts/classification/raw/main/train-classifier.py \
fancyzhx/ag_news davanstrien/lfm25-encoder-agnews-smoke