knowledgator/gliclass-modern-base-v2.0
⭐ GLiClass: Generalist and Lightweight Model for Sequence Classification
This is an efficient zero-shot classifier inspired by GLiNER work. It demonstrates the same performance as a cross-encoder while being more compute-efficient because classification is done at a single forward path.
It can be used for topic classification, sentiment analysis and as a reranker in RAG pipelines.
The model was trained on synthetic and licensed data that allow commercial use and can be used in commercial applications.
This version of the model uses a layer-wise selection of features that enables a better understanding of different levels of language. The backbone model is ModernBERT-base, which effectively processes long sequences.
The model was fine-tuned using a new RL-based approach to classification, with F1 and recall rewards.
How to use:
First of all, you need to install GLiClass library:
pip install gliclass
pip install -U transformers>=4.48.0Than you need to initialize a model and a pipeline:
from gliclass import GLiClassModel, ZeroShotClassificationPipeline
from transformers import AutoTokenizer
model = GLiClassModel.from_pretrained("knowledgator/gliclass-modern-base-v2.0")
tokenizer = AutoTokenizer.from_pretrained("knowledgator/gliclass-modern-base-v2.0", add_prefix_space=True)
pipeline = ZeroShotClassificationPipeline(model, tokenizer, classification_type='multi-label', device='cuda:0')
text = "One day I will see the world!"
labels = ["travel", "dreams", "sport", "science", "politics"]
results = pipeline(text, labels, threshold=0.5)[0] #because we have one text
for result in results:
print(result["label"], "=>", result["score"])If you want to use it for NLI type of tasks, we recommend representing your premise as a text and hypothesis as a label, you can put several hypotheses, but the model works best with a single input hypothesis.
# Initialize model and multi-label pipeline
text = "The cat slept on the windowsill all afternoon"
labels = ["The cat was awake and playing outside."]
results = pipeline(text, labels, threshold=0.0)[0]
print(results)Benchmarks:
Below, you can see the F1 score on several text classification datasets. All tested models were not fine-tuned on those datasets and were tested in a zero-shot setting. | Model | IMDB | AGNEWS | Emotions | |-----------------------------|------|---------|----------| | [gliclass-modern-large-v2.0-init (399 M)](knowledgator/gliclass-modern-large-v2.0-init) | 0.9137 | 0.7357 | 0.4140 | | [gliclass-modern-base-v2.0-init (151 M)](knowledgator/gliclass-modern-base-v2.0-init) | 0.8264 | 0.6637 | 0.2985 | | [gliclass-modern-large-v2.0 (399 M)](knowledgator/gliclass-modern-large-v2.0) | 0.9448 | 0.736 | 0.4970 | | [gliclass-modern-base-v2.0 (151 M)](knowledgator/gliclass-modern-base-v2.0) | 0.9188 | 0.7089 | 0.4250 | | [gliclass-large-v1.0 (438 M)](https://huggingface.co/knowledgator/gliclass-large-v1.0) | 0.9404 | 0.7516 | 0.4874 | | [gliclass-base-v1.0 (186 M)](https://huggingface.co/knowledgator/gliclass-base-v1.0) | 0.8650 | 0.6837 | 0.4749 | | [gliclass-small-v1.0 (144 M)](https://huggingface.co/knowledgator/gliclass-small-v1.0) | 0.8650 | 0.6805 | 0.4664 | | [Bart-large-mnli (407 M)](https://huggingface.co/facebook/bart-large-mnli) | 0.89 | 0.6887 | 0.3765 | | [Deberta-base-v3 (184 M)](https://huggingface.co/cross-encoder/nli-deberta-v3-base) | 0.85 | 0.6455 | 0.5095 | | [Comprehendo (184M)](https://huggingface.co/knowledgator/comprehendit-base) | 0.90 | 0.7982 | 0.5660 | | SetFit BAAI/bge-small-en-v1.5 (33.4M) | 0.86 | 0.5636 | 0.5754 |
Below you can find a comparison with other GLiClass models:
Citation
@misc{stepanov2025gliclassgeneralistlightweightmodel,
title={GLiClass: Generalist Lightweight Model for Sequence Classification Tasks},
author={Ihor Stepanov and Mykhailo Shtopko and Dmytro Vodianytskyi and Oleksandr Lukashov and Alexander Yavorskyi and Mykyta Yaroshenko},
year={2025},
eprint={2508.07662},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2508.07662},
}